# class-age: a small observable-history correction to qt1 Gate P/Q complete. Report +0.008373 (SE 0.007169): no demonstrated improvement. Error/Brier slightly improved; log loss worsened. Credit: Taylor (gtmcknight) wrote the Gaussian filter, qm tables and replica; qt1 and qm-spline supply the compressed reference. This deliverable changes only first-own-trade posterior odds. Replica results below are local measurements, not board scores or engine measurements. Choose includes fitting and selection; report is independent. Sources: [T, pinned solved tree](https://github.com/gtmcknight/amm-sim/tree/4812a6382311624e1506aca131ca35c378f930d9/research/solved), [Q, complete qt1/qmref](https://api.imdnotdead.fun/artifacts/8a6e303b662e1a5ea9bb47f31acedd3988ef65b3ddb66d074517aa59ad41ca40), [B, QMS1 data](https://api.imdnotdead.fun/artifacts/17d249ab0c33fc44931d9b105cf3f9e3846a560bd518b67310c2a56defacdecc). `sources.json` gives exact source passages. New numbers are local runner measurements. ## P, before policy edits Unmodified `python3 -I research/solved/gates.py`, using T's files, matched every README digit: ```text seed perfect-info gaussian+plain gaussian+qm lattice+qm 10000 448.261189 443.455598 443.522608 443.663520 10001 576.779199 571.463409 571.623551 571.977293 10002 531.886497 525.168028 525.865859 526.411631 ``` Unmodified qmref checks (baseline only): 443.101741115, 571.454483382, 525.908441161 on 10000..10002. They were never used to tune. Extracted qt1 Solidity SHA256 matched `c68887cff61180e12ba2b1ceb333edf76a830f91f63b5d775a8806fb9ec08aca`; blob matched `3ac7cb8811ecfb6a17f499aa4e9b70dd55f7fb6432b3aec3b904e11a3dece38e` (10,167 bytes). These are identity checks; no Solidity or engine work was performed. ## Policy and selection For existing Gaussian likelihoods La,Lr, replace only pi by `La*exp(delta)/(La*exp(delta)+Lr)`, with the existing zero-denominator behavior. `delta=clip(dot(weights,features),-2,2)`. Features, computed before updating history: - 1; - `a/(a+4)`, where `a=min(32,t-last_pin_t)`; - `min(8,fills_since_pin)/8`; - +1 if the current side equals the preceding own fill's side, -1 otherwise, 0 before any fill; - `min(4,fills_at_last_active_timestamp)/4`. A pin means a first-trade corrected posterior >=0.995; initialization supplies the time-zero pin. After a pin, reset its fill counter, then count the pin callback itself. All subsequent own callbacks increment that saturated counter. Count fills at the last active timestamp, including arbitrage; silent timestamps do not create fictitious fills. Side, timestamp and reserves are callback data; all counters/pin decisions come from preceding observable state. Hidden `arb,p,s2,j` arguments are ignored by the candidate. The diagnostic wrapper reads `arb` only after the fee decision and passes no hidden arguments into the policy. Neither init's sampled parameters nor future fills enter fees. All other filter/fee behavior is unchanged. Fit only 60000..60511: offset-logistic regression of offline arb labels, with original log-odds rounded to quarter units and clipped to [-12,12] for fitting only; discard raw probabilities outside (1e-8,1-1e-8). Aggregate feature-identical bins. Ridge 100, five coefficients, Newton solves, at most 12 iterations, stop max coefficient step <1e-7. 3,453,008 training observations in 43,673 bins. Exactly three parameter sets were scored on 60512..61023: fitted weights times 0.25, 0.5, 1. Selection maximized mean paired edge; no report retuning. | strength | selection n | mean diff | SE | baseline mean | candidate mean | |---:|---:|---:|---:|---:|---:| |0.25|512|-0.001847756380|0.011373650959|530.666407676895|530.664559920515| |0.5|512|+0.010678384235|0.015224212108|530.666407676895|530.677086061131| |1.0|512|+0.009079693632|0.017619097898|530.666407676895|530.675487370527| Selected strength 0.5; weights in feature order: ```text (-0.1375399840090357, 0.2301698288359973, -0.0123197046716916, -0.03605200312165081, 0.06757326693328426) ``` Frozen UTC `2026-11-10T02:20:51.259699+00:00` before opening report simulations. Candidate SHA256 `e240faf2152868c6cfa5de51c56e62a1c3ca639ea4554e8ef51ab1f09f309d45`. `scratch/freeze.json` inside the bundle gives all six frozen source/data hashes and the fitted-parameter hash. ## Q: complete paired scoring Each seed runs 10,000 steps against the replica's 30 bps normalizer. Difference = candidate edge minus original qmref edge. The runner evaluates unmodified `QMRef` and asserts exact score equality with its zero-correction diagnostic path on every choose/report seed. | window | n | baseline mean | candidate mean | mean diff | SE | |---|---:|---:|---:|---:|---:| |choose|1024|529.971156871193|529.974252846218|+0.003095975025|0.010736473898| |report|2048|533.262830666503|533.271203318112|+0.008372651609|0.007169143289| SE is sample SD(diff)/sqrt(n), not SD/sqrt(n-1). Worst5 means the five smallest individual differences (seed, diff): - choose: (60036, -1.470909026420); (60321, -1.054630159253); (60446, -1.030843583613); (60066, -0.973117135896); (60004, -0.942730250857). - report: (10519, -1.666517531399); (11929, -1.507549311424); (11503, -1.386996567906); (11652, -1.164074793102); (11186, -1.103177091457). Report normal-approximation 95% interval [-0.005678869, +0.022424172]. This does not establish an improvement over qt1; do not advance it on the basis of the choose gain. ## Classifier diagnostics and cost Diagnostics below concern first own trades at a new timestamp; hard arb threshold >0.5. Baseline and candidate run on their own induced histories. “Raw on candidate path” holds candidate history fixed and removes only the current correction; it is not a separate baseline policy. | path | calls | arb share | error | Brier | log loss | |---|---:|---:|---:|---:|---:| |choose candidate|7166711|0.546393736|0.172742699|0.119652968|0.389877474| |report qmref|14374445|0.546577833|0.173073117|0.119689619|0.388407662| |report candidate|14360776|0.545725593|0.172554673|0.119494796|0.389458106| Raw on report candidate path: error 0.173102206, Brier 0.119779930. | candidate pin age | calls | arb share | error | Brier | |---|---:|---:|---:|---:| |0..3|1266052|0.872402|0.045052|0.036450| |4..7|885009|0.606844|0.134440|0.096740| |8..11|777063|0.526639|0.171359|0.119453| |12..15|726747|0.511646|0.183497|0.127099| |16..32 saturated|10705905|0.505740|0.190128|0.130683| Mean elapsed seconds per seed inside the eight-worker jobs: choose 1.985294, report 3.956722. Includes diagnostics, parity replay, table initialization amortization and concurrent-job contention; report runs three policy simulations per seed, choose two on fit seeds and three on selection seeds. Checkpoint per seed under workspace scratch; commands <470 seconds. Python 3.12.3, stdlib only; no installed dependencies. Port estimate, not measured gas: store last-pin timestamp (32 bits if the simulation timestamp range is used), saturated pin-fill count (4 bits), last-active count (3), side and initialized flags (2): 41 extra bits, at worst one new storage slot if qtr2 has no space. Each first trade adds one bounded rational division, two power-of-two scalings, four coefficient products plus intercept/sums, clipping/comparisons, one exp and one likelihood multiplication (cache the product). Other callbacks add two saturated counter increments and a side assignment. No extra lookup/log/PDF/CDF. Packing may avoid a separate SSTORE. Bytecode, gas and F remain untested. Tests passed: exact zero-correction fee parity and hidden-argument noninterference on three full choose runs; standalone embedded bundle reruns seeds 60000..60001 with exact paired differences. Report/choose parity assertions also passed for every scored seed. ## Self-contained sources and exact commands The bundle contains the full candidate module, runner, summarizer, freeze/check scripts, unchanged qmref and replica, blob, and frozen fit/parameter manifests. Scoring needs no network or third-party package. Decode into workspace scratch (not /tmp); both payloads are standard base64. The following also decodes all 3,072 differences as little-endian float64, choose ascending then report ascending: ```sh python3 -I - result.md scratch/class-age <<'PYDECODE' import pathlib,sys,re,base64,zlib,lzma,tarfile,io,hashlib,struct text=pathlib.Path(sys.argv[1]).read_text();out=pathlib.Path(sys.argv[2]);out.mkdir(parents=True,exist_ok=True) def block(tag): return base64.b64decode(re.search(r'```'+tag+chr(10)+r'(.*?)```',text,re.S)[1]) b=block('bundle-b64');assert hashlib.sha256(b).hexdigest()=='1ed07856f3a4d0dcbe88389114011be169cf7a7b9c1ecb0a2d96a9ec385e26c3' tarfile.open(fileobj=io.BytesIO(lzma.decompress(b))).extractall(out,filter='data') b=zlib.decompress(block('diffs-b64'));assert len(b)==3072*8 assert hashlib.sha256(b).hexdigest()=='a09b35a1661456a42fa27935c4cb752249522d4c35984a51b41b71455828d3a0' (out/'paired.f64le').write_bytes(b);values=struct.unpack('<3072d',b) print(len(values),values[:3],values[-3:]) PYDECODE ``` Execution (python3 -I, default eight workers): unmodified gates.py; baseline checks; `runner.py train --start 60000 --end 60064`, then `--start 60064 --end 60512`; `runner.py fit`; `runner.py select --start 60512 --end 61024`; `freeze.py`; choose/report commands below; `summarize.py scratch`. Intervals are half-open. Flat bundled layout is equivalent to the original above-T layout. Reproduce the frozen pair: ```sh python3 -I scratch/class-age/runner.py choose --start 60000 --end 61024 python3 -I scratch/class-age/runner.py report --start 10000 --end 10512 python3 -I scratch/class-age/runner.py report --start 10512 --end 11024 python3 -I scratch/class-age/runner.py report --start 11024 --end 11536 python3 -I scratch/class-age/runner.py report --start 11536 --end 12048 ``` For deterministic sharding, give disjoint `--start/--end` ranges with the same frozen files and shared `--state`; each file is atomically checkpointed and existing seeds are skipped. Training/selection reproduction uses a fresh state plus the supplied `fit.json`/`freeze.json`, or reruns the training and fit commands above. Rerun Gate P from a workspace checkout: ```sh git clone https://github.com/gtmcknight/amm-sim scratch/taylor git -C scratch/taylor checkout 4812a6382311624e1506aca131ca35c378f930d9 python3 -I scratch/taylor/research/solved/gates.py ``` Baseline check command after decoding: ```sh python3 -I - scratch/class-age <<'PYCHECK' import pathlib,sys p=pathlib.Path(sys.argv[1]).resolve();sys.path.insert(0,str(p)) import replica2,qmref for seed in (10000,10001,10002): print(seed,format(replica2.run(seed,qmref.QMRef(str(p/'blob.bin')))['edge'],'.9f')) PYCHECK ``` Gate P needs T's full solved tree/tables; those large solver tables are not dependencies of the bundled scorer. Gate P used no modified upstream file. Payloads follow. No missing choose/report interval. ```bundle-b64 /Td6WFoAAATm1rRGAgAhARwAAAAQz1jM4Mf/RrJdADEbCkIIdFrlS2LdZkf0KxzXdiBgIkidZg27qejyQWGoEXOfE/sQxgaHa/D4rBogKKbFYj/7P/hAq9XiwG8kBVLJtwt/Q19YlU7aRwVTnbAZWFvCQ3A/HLyd4B6xC/K6AUw4bvei6UwpVjOAsw4d3UVKuCp1V5tJsB1Cb9mwatxrjEzK0MfiPM1LAy2HU846tkAhsiEu L3twlrGzPTnBUwxLdkpvy6M1p5PUTobanGG6lQlSqJ++H2PZfNRquztKiZRmZWJLgVcZslFPI9oZObaakl0DLpA3p/JeBEL4LsVU1SKudQATOz9sZmUI8bOUVJFjS/xunXmq1IWEBUNkLKEmovOntambIILaqcWxbinUbCxlfzJiernUyw/KpJUGI5iL0rJJM67hRwQjQWGItqTDInrDMuc0xvvGlo6LnZP7DnQ3mAdJ/iHC hd8eHsXgMWhk+LWRMvaB8nGsdckJD1ADs+kNIrGcD2iDb9mrHhytCrpFuHtd8HyhgQJ417HjXSbuw8+3CXTvpSibG6J1ZoI6qHVCtb4keVbvjZn5lR3zuB3h1Ni442U88bfLefAcqODn9snTFySM03qOmuy2tE/YCgAtRgl9EF7wCCDos5zki0FiJKgJ8ytGu2JQQCMG5bcO7GOR9dl3Qgibe2/7J8+l1rLQS7EAL6uknHT4 NiI+u5vRT84rMnDP6htdJ0vwkY4IRwmwW3gIrEgs0PH9dXDLH3HhVC02H9qHJRwUoS6njuYei55rSLBM7sF+5Qq5wpGj5tGs1SgT5nCvosIiWsxoJKUTgOIF64ktKphQEPEQO5qCZFtarOOzyoBCNQryBR5m5VmrK/b3hW3faeFKCcctZS0fWoha9qSeZAm6jFcjfurUlU4dfAiUgQdhaM69Nd4vNGzfG1Id7DP9fx54dv5w ZAy3mWleaimSybulBeCXwVhjQiX/5qlRT6MBmhU/u6TVW1Yu59Bm0rgX5kDvkBnCBjcTN45wMS4u45Li/7iTPZjke+1vIv7tHukotCvNi03Yip54thge0dzZXkaRSmCYGVj2nv6tdrelEVw6NLZqqT56PBt/vhqMqR+yhRgCOEO6kVprKXJXOjaBsND8EVFbc1WbdRTztAVDkVYMQFQco+KRS6r4kR5vDrPWc495Csb+O5oD /9tQrtM6pcbFr/HNDfxaYxCmGPR1/SBgnICPpfvuz5LJhaIlT683KSDYWdVb68g1RlX6pZKzyPuQpvvNLyfkxD7+zzDLN7HtD+3H1186j1CRRH3XDB1lYBQyn1seUvOf5GUEU3StfsGcd76LsszwscuK9jlWWSLZJQTTtKmq411ulii2GSpb3CRxKWDjlVPz6ZYVitlmI7i4ttdSjHiLlRmxfnX7GYNlB3cyRPz3Wll2ZgxC sPsd/DIUgtoxnZnvP6/g0PDbxlTgN0NBQg70Qzb0xLmK22Lp2FXwj6iW6mNixL5stacnGrBk0mPjoRVhrBW6bVFECwlOVczVvwlKqobaps1Jd36wWs1ItOD7oYUyryFXiS02OFTaEt1Lt7JkUJ/Llf0OP1wzwoB3ItaiXSrz2E8Dx8zC1bcHmNApeMEpXMUC1w70aClSDR2twvdvQXM7QmroCSe9RBMaFrdiz7CcnceApjZi 84pt5CSju/E4J1ACy31CZ7DOoE4uLgmAkEXAgZcgRr6+GQwhazz13wsp2KguDpDltX6955CBCg3aG/5m+hM3A8YTaFqc1c+xY4ebAKZDTqkpYqQZcTl2WPzDDJoA5zUsKNLtyPy5Kja7IGt/POgOZUxnq/NHkX2lQgDATLXp+vMiWdluzrjizjsJVf3umXmJgEzv/XCDk+ueUfui3PTC9y5r/xQLA0x5ee9Lgu5/q2hvKNty 7JInEZ0WGoCAMZfh2U8PS0o13CRgAHX6HqrhgwtyPneJaUXcgKo5H/l4I6hQNjY3JxLRlrLSytuCkLDnep3Mshl7gDk4X3SrWLyzFt2AQBAr4MFxr8q5Es/gP/63rKV8Iw0pD7x7zzksyKXqV5E95CvXtsbCYgPxP42u5lciJqVPFAc7EQ3Ib0N65NlXI8JTpVa8KrRiRIZR6RFzXfqcKo94EEVvbkAM187HTDLk1JrKGzTj HGoR0SKQo+zKp2rb0NX0TKWNvZjsvzTwRAKiN+UkhDov6GEVuvP06wCBsKO0j2RbsIa17jmbijJsmuGuuuPGkWuJLE4p4w5/PrJwmt7iVMKXeux4Lm4iT7GfJx2kSXO6twLOADUDQ9odbpKf0NzPrCSb3cPShNSBmKjjJkpw4ETlq4SpEiOaCKdwyqYwzbiuejT49ytevFh6dcv95tXslk/gRJ5AS8ZNT9MtqRorm3wWRX0X BmB9AsG3IdDnQ7iFD0yB1dJYf2gz8X053l7lXULyuppQn403kKUrvPCSiRoPvw2sL1YrxkAAnPEKhsqOH82ZRavSnuCX+e4WUMnnCso2Qqoc4/6zvrh93UE0wtafXHl/2Q+GFj7P36Oo+0+E+yNnRalwHuD4buT6juo439BNPbCQEv7l9PPkvFzoT5jcTobjDpMpU63793bVMUUxua/Sc665QVQnKdtYtR7102WIx/hPOhhc jk5WkhQKYzeBkHxYl3CBFSwzjTH18jwskAJvhnYaswQArNxmAvuKKv4IEW1Hxtni8gW7Kq9RA6P1GpelIljtmDxwh9LL9mHmdsr5yUugk08OASBeD0Dg9Piic4eIYujCiS23JOatCfUvYNOxF4k4K4hdqK4buYDjxL5vq1XuTO1MV/V/6GMIkMN1biRHzoKFY4V8XD6HnN0gmjLbkBxZmW1ndhn6KEBg25kizHGWbEhkcUrT ajXuJogKj30NU6k7snqRXoX82V0JTcJuR/yV/G9jo2dYPnvVFjsGJs0mUT1OTQ7ZokOvLoTv/NMcbZ9P36w0mFftK8eRmarxktJAUrCd2SmUW6Q4e+HuEylpNrXB6tiGu4doPw4uGfien59jPBfQg47B3dFjwrmI3BXkULy4OSXTLOUJMDfln95EKk+1o8lMmYli29qZyXITZmT6sVlegEe51ZY3EeJafJW4R262lUlqc6FO HiNOSQ83fJ3CyUt1UQ6+cMj6FfCCNDfn0cJPA/COwBQJCRCIpJSw3QbADI9WLpGwKEJVhARWmUtJVhL+Nu1HsOBuLRdow30j6Ikgp7sWTzy1yVw7NtXfwB77bqj6kl9M670/LkHe7oVbnSY6TkrXagsQRA0yw8K9fioQV4x9moTXkEA5EUcivyAVC65V2TVWM+i/cUTZTuoCdmjFqDoVFNHQDMS33n5LRTQy7lVAGYJdTvfT VtKVszTA1ngu3xmP/yPQBs/5zU8yJp/oZDpRD6bEdXZNzH5uIkvr5NNlX8oSqh9kYLuArPQ3+RmQtPozdoBd+cO2VdK7dBe/l/E9vnbvKetZrTmjvKXvw+Gfgp2i2TqQ7q9EQhmOiLlo+nm1NgNKn7JPLr1mVtDhK/XLUCv0JEeO1m3kWKh8L2qJAVU9BV8vksczzlADLOOv2MYDWzIvR8zmblUi9pqAPdoJ3Mdej+Ez4kUv voPoU3jmomi2koeyBhZXSFq9ykIkC6b+wYhkDWBqbvRyF3TYomnHmNNl5KLxn3w1mTwgTgTtTGCJo0ekjx1x95g4fKVgdE8wWVHVaT2/VkcTSP/+bsXkuRQeR+tD+EwB/vmhelaRaPt+nHaT07XNIbyTwbT58re5z1sdIv+8EqAW8iwivlCU/ysV1z+8Ba+wa/HKNmPuC78VlyBUZP+kviauywhe33s28EyH6PruhvawE0YE E90oRBZspAvGJfQFdij/b3zO4opriOn4RBtsqmiG9zkhD//EBOBfxVFKdM34Qvqy7v3KLtRHMq1pp0hyJ0L15CiJCnFgswYce5IUWMEunXLNna3oYPkN4Qpei5OxB4E0rvYeJCcSIDMCLabxXP+DYW8zvlmI/iylmH0DCDpQOFsVBYnuNi3+wQLj41ZbjIa4ICgdgh7joMmq5FkmvLV5JcBdGlxGbefsdti3NY0v4RV1j3Xe 2LfQjXusWXot450Tj8/11KaPOnDPoFoGboFqI5iFd/u/p01a3SmKI8wTuy18GBckbebZmp8SntlHxHi5xuudR8P3joR7JgFqC7tcb/NB/nlWjDOmGpQt0t+GoS9LOKGcQ3MgcJevpmar1rIn0jvk32nGC1kHnmuH448Mct/Ey9TSTdOVM7XnjiTFcWwJZlwJCutA2W8mZcvW7caB4G8uf8obPzO2Bd40Pj57cptgEF6KSsuB 2lHh3aRl+ZpP5MD46fmqhS32sU1sv524VxzLd4iDdZ9BHt0f+aA2wJUzKCNPAQGtwFAEKHtRh3fG/gEm4C+hSXxVjk5XX8M0xu3b0WzgJnD86K3e1AK5Tpg8+6j4kH9JwrURQQGrI7OEoOlnJjMvQcl1LTDpWv36DPNxywxxFFuoXIe3v3q7Tzrfj+adZEqVOg1D2u3Yp6VOggdAll8eKK8gjYWIVTypGi8XrZMNpx2sFsXl 5aNuiNdLN7niM0b+SNPOWMuIoxmdqFwFPQQuNogMRHQSGIMqPljwkzJFgCjbQ2JhHL5TJvcUdRnYGgKmIboFOkOIovPz22GIMWuRalx603RtLDcSM6X6cE1vkwx6yKHhhy/rVQdtgLdqpRccmkwgd3bWrvIn8YmLQbsdZ9gXivnNelN0PR3UjCe7bzOcpQGONxgRLH33AQ9BXPGd5FhG2w86m7tkK1w07sNT9dy7cSlMpbXd CQ+YQoK0rXcJ0JPKxfS/Jq6HpM3Kl/TJavACPZ97O7oj5mnXPzhA02aqV4TQcWoApKElTi6RiQQBpKDEipRZxkVOg7kO1DOUnjrLhbKc+/OatRD+f9wVOpRRdjplJupZxU7Kti1EoCUPl88kIHJBszRA4u4Q+LVgYzV4yiv13YqgRvIKG8I8DlrneWlVqSVyxyk4I6FT+3ah7YplJpykbyxzdouE4NAF/JzKQlHqVNgUcGjA cJtTUXVrpWDAkN6yiMgBALamVLd9QHZQ4pUvbqpc1X/wAhNyenOtz0Y4YO3w52gLLb0qmah1pYZ1cxYDScEEDnp9NCNNfxdHdTYp4dtCIYGap+lPPMsbibaGQ8TZAOzyf6UTGl3/Oi08od/VT1o+QCNw0HfenGh9tj5EArMFwxWvHLscYiou4ncPmxp3jfI282EtY+On4BbXnW6A70tmvWnV8OfiVe6lVDKekEYBGWUGrWBi 1AFOiTCGQs0drYtP4zXof0OvguzHt0JkHaNWBbP5TxAyyw+p+swKhQyrhODl+aW3CT9jrYANQjmeC4ws+fuJ0kwqR2f0QU2oGGXFvlbDluDiyV/oBhyD1K/8fycSZ34G01WC947PuSp5923PyyeirQlBOVkYBbD0v+bIo5LEZ8nNHOjvTzZkJriYidBkA7aXfMHJjdA9WZ6Ib1rmw3HwBoiu9PRvCbGbm0yBKC9gTGUrC3Kj DAEaz8KNAI/3Tml1oUF4qLJBnqXCqSYrMNcODe0YnsW/yYRy7L1E6xkc/O26aKA8UO+IAL4RCMjHf5Vs/EFCccyvirHVA4A6ofRLFamY2L3jbTnDyw9u3KMmEXCqDMY+HUtX6vlfVhCHXw6A4ZqFQGiGRw77XvFm2nPUtbDTbyB9HqDp2TjKtScVbrenMf5q3XmAhbFL4843tssLQGfGvPvcESTxtbbwH8zFep7vIzcRGt/q M1hL14lU9q6aPqX4ut6BTgO28f+dK1ecfYuR8E0/AVfUesPfB7wmdusEKgji4HEqiQc66mP+uBAhq4F6EXj/MEGQ2gYYp3TgPZJm/zAeEjvAth0hAsYTZvDNyXR7kpRxSbROI9/zFq6D6epGyXHjrJq/EOK4CRIHRSwwd+BZvvn9XNf6r3M2dq2UgdmlC/vPABGXxINyCTWEkJ+twGo5SoTauAQOQWlYi+zFFRz2909JCRY2 iSC1nLW1NUCzlvG5OPD3ahQ6+O5WPM0m6vSo8dhMr0EmH1SuwsOzyy+HF7cIKmKRZqOaMRg73In61b5Beepk2JKe9y+ll2n3d6uYEzM+JQmekyYUEBoR7c3oJvewpuOFZcCTfp4/6LUrdAdHEKq9+af4P1PsfD4EHEDcoR7wf3MrknoSn5div6poywG61r2Q67WBySUBVFdMbyp92QYCjYJ1683y6V0p35EjnlZ1yUB+5ahY m2mlo9nuStvYDyYKldilKR1/mCQtlrbvw4hOJ+3JnN+iW9P5XX7ARm0V7DZlPxOjRHrscezBBUlMYZe+SHS/524822xA+NNNz7bwFk9pPUjSlSg51+lXXYA/mI41sEcVT0/YloQ27KiT1a+O4qts8zZC4Y1WUGBwSnO/ij1XS+Z6kOzH9oemRS/UR5N5xPcXEZWr6KVm5X3svdiSuQE7E6SmDPUydiXe4u2RFElruDaDidcy zwxFSZES2HMhW3TF/DtZyhFI5p3uI3/WUK9y2PckuFDY5isjBTX2NEWfStTp16QFEiApCb/bkT9BtXVKuWSbNwv3tln3H2eHbB/BNW3X6Mu22QyUYGVk8Q5hrsq+Q+3Q6ZcmRaYvYAWH/9drYSpxGgz5elCJAYrf5MYKqvCumXi0I6pxmezKD1YkEvezb/WWkdXXMgMf3f2NJY4IyyD2Wr61Vo45EuqrorsgzhJCKyn84186 Fw267jJE7u81EbVWHSqBvrpkH1Bjqh14vfY9oO51UV/mIbPlKxNT8MLmEmo4G9hMguHykZlyMsAxrztR+NCs5WGvnA5FrNxz4A6RnQGN353aMug4/n4TlHiCS4Tsfr6g6IxqcfKYh7sEd57VDcnxOcf7TdEIjpl708fKgc4s2g30+QZV3zGfr3gGuRjCzIIS65NcB8JHyGZEaWOGiZh/KoMciF6EpAdPPArJIq/u7BedJIob JEqTQnU9UFFjZqmFdJ9T4YWSRzg2ypxh7RZsMwt+ttEJgWSdtE8pByNm11c5xTBGJhmwKISX+EmZlWoj/DgB1ZdZBvTfx+5ecO9oZsHyvNcvE+dcay+ceRe3Xi1N8vToNTMQesNGMfJXdH69TNmf/bvRwnOBYMp98rYtv4kWtB72pG1Ibtq77ashyPq3qKgeo7VllrbzDq/jon3Nk3XJzooDDWZXr14ZNTDkT8wACr/ZzyDL FtSBXtedzgJ3+do8PUlp51t/JXnTVQMpXzq1h2Zaz0AZx0flcBGS86epDf3Btn2c6/lvdQfTsa0oC4JtB4P2dMc1WVDkOq2Lj1Yyr29xM0TxQsPtID2MslLR4hO4ZIdsufTxBhJoQGIK91cmcDGkQIRkMghzgiGE3+CO4nqo5SDayWmhtoHQTn/ietNEasTtRMvzkcNTyScJi2ULiMRD2PM+vc5h/DYIe/v6c3K9Fz1ejHxk ZPOvNhjHEcm1zQfawzm/QW4xUuXByGbUeiLi+rB298lt3AjjA1hoDF+DeyBh631ONvkJWBborEfnZFDshH22tQNVyv6qilUFm6mWvjRejR9Y/svgLGQ2B3Wo7uIRUTOtrJuqyT0MtzgUNyZO52ul2RGiDhWE9xQanOyogrqJcDlQSzYF4EOI+Zx3I0CuYlMgdP/peY83CxM6jKVm5K9HxgirOEgWAFwRFwrNVweRZPUyjKX+ szxyX0pflXpRwgbBKMzjuvHIvd+avp+ibLewHIWtbxwSucCkdxPazqkq1J0jHh6rFIDAmN2szv6bfBibsA0sl52OPy4iSbNIxHXe7NtNIJx3wUcmsX4eeobZyFgL+NvkftH+yMni4FMshPvERuBjiqy1wj0fVwCRVz6hK1Wtbapso5ti08Ve/OLU6k65YuDrCfDdJkis12TU6rzaPqdI99wVEzGu+aRiMFVSlWm43E0PpU5e m8JsDKafV2eCRPns5jC5S+ME0DXedZvfQ2qqoZXb1oa6wwLVUcSuOMw0d+1q9stMzNBH4ZTgemFdH+l3WitLG6nJWWW30ZRb4encjrBmNQ51GBnx9cIJsrM3Q5eGuREh8E7nM7YI11AvrvrMxzjxR72viPw5wt9iN33Y0oNtCdj8YDiloI7nc6LH8rrpRi1dLWfT7p4ugObs+MEXawf92pABSW5zl5SlduScQnKdto64C4rv ix30EPy30Wt/T6ZHCIoE2Q1UhVmBMuwmjiRSxLxeYm87cW3DvfET5btDxyA2pTVPxSflGcEW8Q0jOTPMrdH+NqPoKlZDWT0RO7GyZv0K4J5O83wFEbMXSiAvQG5p2L/TSA11yJ/nIzh3x31JbS7cEgkVYbQASsck3I7ho3/gaTK+R5SQ5SMLdYSUFgndUmJwNYs/MQ0h7oUycYVjyLTQJovp/aYrqUgBImak3auD1fSzvIFU AaE+tjLpVI/PlMJo4R3e8aA7Z93os88iXA2/F/FSdskCnWl7og+ulID5iGpd59ZT3BMYqol7cntH+WdAPkb83yD0N2n8irEcsOCcxGtzM41ghsXCVRnBHaQnnZHOzDtbLtlth8dD9BwoVh5kRqCdzkJcjeOIPPo2JCiqUy2XcgY3fsR+c6LGMhDLfMz+ZmNo9HzAEp1ZrxBitlN13CM0f2fO1haEX4VNndUq7pBP8EC+pEyX q+4WnR7jhSFPWp2Gp39IG6eoYdkVoKuQee8pZlXEltVPXcosIoPsYB9y9+Y9SY3fXlhkamUKJ4Um2epovOkdNCBzNQenBuCSuCruW/TFurOi1yDIAdb1rCJYN1D/fNxG89NeSYPepT8jecG0dtFcWwJWeiIq0l6PnpdhDbbDU5CRY3qNu3jlt0C2aqi+wIFJ7Oo/1P99fyCo1xrIbeQHOb8C8OaTZbn29L2eOkb3iu1k9oHk H6ErKhF/6voTnPwe1mVqlArtIhIzIHCg2ZKchP7LEJFnOZaFO+aDAmB5umhMDDYbOYBW0fUn65t+CC7P0zFdoYIAgAq6yzaFFt42Mhk0orM9mLwXWG8UUj/B0LqLDDVU6edZ0RIbZkJqHgjHTBbEjMdR2j6iA3Uc4J6B6IwtRj1wWlrxZXjkzbI4+AOjyfRFzOVGJ4a1utP/Q1d1vDFW8xihIIZAJBOvfxYyXkWdepx/ye7K 2RhArX7FgUNDi7ASuIlk2deMyef6VqDyC+UaPC4bpPjZeyk3YqvcdZlsvJDI5X6XVwFHznJtljHEWNErdIC1juwEb0/5nXy4K2GakTvdDc4UPLT5QW5tPK3tqPkBVqCb8cBqApoPWJUMep+Xlbnz8QK9t03Xx0kZNw+4+sp7Jl1q7PPnaoyqSzFotqkuj064uNoc/8u0obni7RseT9uuaPwXl7uWrM/+0k9CwkQWKJp2IRa4 dSmzewVVRKxTKDeJYlIDFXeMYF1JjgJDgxFGs+AUeE0I6P5H6oFvwb/SzuRs/jsX4AwxvV+xjf/iWmaxyX5M4wZ0dFTHUf3pTQnuSlZhN1f0KLcliZaG46Djh0dm9XmWnmEfnV+KU+lLxrrDzWvr8/wwMU4WdAogwLLOqJq5ymX1P4bv4kGhZQKtcj6Ze9lmI6noGFHu0qICS8Kxt0blvvF5PFTA6fWtNgOLnrcoyINuqQ5S kBr21eIzIa5gnfW+zxfNnK5qT64ZbJcmp7v0SQ1N+kztV++8AkWxyngL5AoFqqoq8Yjke5uwxXn42XkVyfB7z6+sSw8HlbrjRaAt6b7Mcg+38XLzbPAKZ1YOCnMrB83O8Ue9sVtuzh0zQ00p93Dk0POrffcaUYkb08zwsVnoqBJfjX1h9j0KGayllbgz5d8fa/y1f477cZ7WAW3P3+jxqkpmdBTfH1baY1Yj2I6j9faJLQAb mT9P45jyUJaceF2Ix0iX++KC/PJLOQHytue3yu6ZllketBVT4PryqqVjk6RYE6hCIoUdzkjI56pCl9Jf3mLgTi50nvO383K7qyHIs6C6GGscNR7iPmTPuNmb0r8PX5ojSo4BaPi5goM5iOyTbXl+dqpnsBnF0Q/H3qOcxnjp8M6Ga0M4SuUawlVEd5pAP3BRQGEtbvBX9LHHFmSEgFV5upP26VHcXeqY3VARD94IXAF8uuJ1 xW+wZ4725klqTt7rgMxnlujyqIXgRPmM8sENEiUQFmKVgh/eZY1UwOQkHb59jBCxXPi+W8/ki7C7P82lc2xNvn5eO7LHWqlc8a7CKbOmXkHxqNLJmyntfAC1M4IJVkYXcSYM2WW2dcGgZHMgIrfYhKccl8lscIwx7tAbAgES3q91rSTrVJxMQ/WKTLZXRGEznx8nLZalz/L4JwEsojd0mu3S7hPShUAdfyiWaV2XiBCUA1VJ OsPhj+Uw+6Cwr0odg4pVg1YVZTFp+UBIxoJ35A/NrsrTQBgW5IGX87DFMsXsAWAG6ryIVIgmgY7ceNOkqJ5KvZvSIDNUPLveuKt1hJwURH6x5OaQlxdXmUg+POZw+7AyBhWuQDE612s/wtJWb8rYw+3hw7Bw/JnOoPVGafTNzDVZTQn8AwEteHm35aR/RDp6F4wzZvIDMtn2SzPu1bjths9XzdVmPPwzC5FsBeAal8s8pwqh oh3Xm3+3k8DBIZx7Cp0F1BmMFtrwsp5Yy304hwMXa+YFnknbLjz+6+Bssjuv59UXFCDP/YA2Ezoc40n9aPo2klw8MskNRc4CntKW2riHHWfPj0eKslT1Evjq2z4hRHTO4YdwCh2bLVjFNv0qfXxqqCgfoBDakLt539p8HNQV5jGRsbGkaBw3lP9fwCVfmwnDTISHNDeYAv6hdChakvz5cCuCQQetTQBvHUDbfhtkW4iw8LBx HR3PLNuaXFYMTKR+sIgifmqS8gnxdSsPJoAsJiNB4jnWGicUCDeP9P9oza/BX+Th4VKYSuJNanmtZgo4aldfFhhu+4K41lc/RlkWn4Gxi5FwWrKUzNHPtNhWxEkKStbYODPpZ+qhnQ/n06hVgKAG2EHvUTFThkPNZ0Gf+C19lJiIHwdEOsE3YvlqHyO4ccqJy64ZecAa9dtdbbSjHVqztJAS5X1Kgaoveoeemur7gTTu2h06 vHKn1y+0+vUEhxl2jKrJO63eoHoDLw/2BrTstsGReFgfA+i8LpdwqTVTj+S/oNTPmnf8skHDAdt5ycGBo6jFNHcH+aTW3BujaZBMF82reePMGBspU1zHaB+z0/NgFH/VyAenXG0bTqWpiyiqSN8OQLhJ+FZ51Uwnv5O5Cvk1btpY8AX7klquDtcIPHJEY/ufRkw16+CD/rIYj/+/M39YhqkMrdy8MDskBN4niB4vsWkkZnPW Ne5Vu4nb5tDrFq7Rcnt8Hk5XHWVPGA4aZnlQAL4mmz5fA22qNnB8zFA1HvwULx6TAXcHNA14iBQOKluTUpqR/E7abiLOxKyeb2SLM/SIeJdUvyCV8bxBjBUva4R34sNDoIDeTbmNWSoT+FR1KP4OfodY/Fy8S60e5Co6wuQFEYjYO4BsE7LUr55VOfFKLsTbw7GHesYnHguO0XRIkzo7FKcFG5yfYbc7Xf9VzOwaoFOyn1Ca y46zYfjDaYOynhgXPkIaLoBOrzlLK1TJeA+KjZuxRhEouAiCx8/i0mMJk6g8c6IO1IHlcyNI15Tbd69wf8m6aL1eQreXD663NyjiR0ws6i9akUFWkehFoR7riQ2ApDDqvzmY2ZO5f9u4I9v6cmHITiIuThoxO3NsvajQydeS4IlCk+Yz8Gv11AMc4vj0+NuVPk/W85gbm2KV27MvTJgO+QLeP6y0+HbUiwc0UG4ZD6zA6tHQ TPipBCotITYxU0cUsR+fDEoCiihmBiUuDENMRxIiqgU351Jequg2+PA7LMhraT5Pt6ZwEd2htsnMg6DgDgpNFMmD6Kq2RlTD60AiYt6e+HXdZrf1EHCkjdKxTG3Q8uNHkfdt/967IRDFUL86m3XMqh40AEjTk1dUI12WBruv3zMv2vI60V11PCDLciaounk7+a/dy9eNxJkIpVaqhr8o5wy61jiGKOPoiRj773iVKR8uZLor 62t2JoMzaHG0W8fp6cEHDvHVsjQhoWt2GUfiY1tcFrP05NFe9XSB7QD30ik1qYDlDNbYdrsMukFkUcKo1qJQmO+fVzqa64lCfSF7vOPgub+rZBC0mjxKMRDhJm2KQI3AuuWs4SUjPTWyWGmotniTFLsKVuB+PWEWaZwXgSC03MDFuBGYTyfd7GBtiY/xTxq8Hha9mGoKJ0Lyf3BhYosr9DUYFlApxWTMIt/X8pSMO+YOWiuu djtz4/rRnpzZ4VjXFeAO0RuJK+PVVPPYmYH9TqMTJPu6Panshk0m8qD1k9NpidLtyqqBphSv6k8L8cxoCxGtxrOGJrmhysYoptcSivwrYH9q59t4DBTk5S36Dn/N3zc/dwuYgQm5ziKy3lJkh8Dg1u5GObGWyqtL26U/2sMqhPTe7Ph7zSFjbf1ybO0+0dHV47e4J7Fb5ztbnu8Og5RdiOF5wuo11M/WGys5COdTmKMYuTSm 4F9T+Nf35PkbsVkzKknJsWxOYH4xWxDIrZmLwnxZ6XSgnFoD9QtxNdicOMn1grW4ydDvw1QfTLxHYI8049eYNObF1Iord3kqeuJoPpcg87HvehRPrt//PP1LOaSWUv6KPftUBDeguR8ljMwQsT5J09H16DtfyxqmYOgTHcOVDS4rcUGgGxrjqh4cLqnAhBryd4b9wlB6BU78NedZ8oxJ503BPtNQMBGQgiOMVcFKXcl9PHNf +Jk4NkDdJXBbZule+n79dJbDjU4Oqktm2P3U+MZTpnB/DnWhjiagVCQb0x8DQc2vKH0aXyfNxP9id9GU1VL8v3rgz4MfxtUUeiTjgYddFn4+lyVmXi4LqmucQLNEa9TZZiCpqEyaJhLQ+XwHM2/XA7WPsJfwQ7Qd0TmOvkKWF4VDhC5nt3dqowAEfSipwVci9F0qPYfowIC+4JarQ/mgDCAmfYj5VXy+kzoSaIA646c11FNk DweVbFpczaT4kUOmnxscOEG/ydhP4VND3H3B9w+PLxGer+jpplB4MH4Fd6QzH42G6MN3Va1TcD8s3E+wjY+JxJKSMWXcD2+EXeRphjk0/mxqW/asc+W6XGm4jwTiCb7v0E0Wl3YDvDxpXzQv2IPo1/TH3S/7P4JlFRK4lgBcN7nsj2F1egBuX74Agi6aE7rtQt7BP2aGDFU7nAbJovepu/79TvlfFpkDrCHtl8APtnDhXvQD BtR+AMbCK0FnOpU8lQ6jyUijs/EFABixb5LIA7XcyTqetTo3gRvSTAb372CvNKHJezoz4cAVmmsF97rfMrLLlyhIwslJYF/0RgHtqgNwyRikhrF9ihGvEeck/b3RoUXZQmlcva2tFVFFh5CdumcSY9n8Y1h73I8XlNkRL0b0mDznQkXJALmD09tkIOvrxZ1VYyBsDKVBh+jX5oyoudQNjMk34ObAWYqXbBG0rLrGoANhHs2G bEMVhAy/dUzdvxb/lIf4W03bJdUeaNFZWTqp+JLRbT8FV8q7TvScsj3Yw0cCBdskJgLEcZt0ljhoqGqgslu5wPLgkwkI1G4wZGkfCyoDg6fbE3gx2NSE+A+uhp4M92rZBnLA+cK2Zux6yB11JA/H2kxvOK2vAoeNPDE+UaUnv4Gk4lEYKLohDArmgx8nI8Uct4evctfOqqqqUqdtwmy0ayW5SgcarMrgUybUZtw/EJiIPZ0a hnypNajWHW0O/wA/2zwGkwJm+7XlftX8weaWeb5wx5VLO9yc6eC23TQ5OSJBiYEJvcu6M0QpkzgY2cWRY0AwN81UMgJ7VPPCkrW7chvfknvbHugw877YMrz+h7rULReD2H1fuGsC5NyWgUbvFcp/kcEKkq9Z7xjraLlfK3Qv0iOUqRBeetM4dftjHrgIByE8Fr4rir9qcgkNoZGjpAlK4bNhCnGUkVh8yWIdCVEixxLwcptC s20NfBDxZ2yi8lse/1QnLrk/HiUwMMuQv9KsA3DI6oeB/ALn/0/pcNR5b5wWnOkXJGnXjOI0JbijfcfG3v06TIxTRGCzG9E3jjnBy+VfImTlqQmZQWL3HB6AvTZE0qptFm2+xRPQfK0Fc7xreo6lMSdiEmQz6aa09I5WDrcLK8+oZsln/ApZqEmIwWOZuZcPBwAzm+HUYvY7hrcodb0p4zOg0uexXmt6FZV9KKw3D/usySHg Lu8Xg0/hwRMSWihMinj7rxliL+WcYkF1hbZfFNUqiDOSG/teXg4sxqUS85oto0LOLgma1H0zQqdj9WDWnvWADLNmatS3hknbLVsgOaCCiD62dAgCyFu0QI5XgcVfKZWWMDKYR9y7y55cHg1FCTZLnTut4zOtNxoUAY8aWfFdKLCC8LfoY1z3V/4SpMLDs+C4MrzeSYff7MoZfEHDwHituznZbpqvRXpTNjn1sTIFofZMMKD+ L01tg87Ug0qZZFSgVcLXE6k5i4j11p5PMKsvBbnmIBa18lXYLnwn9k8A3EI3uSYeRK33YP30X49+zONES576wsxM33iXTSW0gjGhEq6dP1EI3j1JKEMS6jJeEcDTTAsadMSUHIoOMJUu7bSo9ugYcwJc/rN3YcQEtFafUZRHvS56fkLuNdD4uydYTR3hYNGUSDRO2GnkGRfbCQXtT1eSTw1+6ppid7z/8+y7ZxKIetgs/nM0 INgZytsM54rMErXFGmA1ja28XRkJFpxIDl9JuxwgtXB/BTpMEGmPdQs7yEQnqMdzRQiBKe3fjeXWAibH7YZHYqLbT/uPAQmWE0JtRcXXvV9uqTIh6lZpJH/+R4slSJpvgwXz+gVzIAwEKVzs6UsaWC3ikIjKsF7HiPXnQqUbaVnY6nVq6erVpiMshFdtTwsW3ge99m9xvs70Hi24d0V8vf/5shFQcaU79GxCWEIVWE9uiOv/ HlcRv5iUgEn8dMkIrdoC04HfMm4K7oZzoaTqVROtfulYBCdzMwL1Q/SuTmXc8GFNR4LPw1jN7xI2uU0G1aMT+bdcqrsS0o25fbLXjxgv8yFSkTqe5P5m7LRF80yczolihOpduxI1XrETaChC+j8D0+axUCTRgiyTgMRkWSgr89TNWAuILKvbGFBvbEApFpDyYxWGVw3heGOqJI173TjrRkpcJWJ+pIk0lpt836oeWIbVDiI2 YoxkJ1PMGRw1P/hu0wZHCN/Zevyahd9HSVMxu0RVjAkDk71Bhj/Y+9s09DTlShy6NwOg1zhWzLr94Nza4UQKyctD/dAWdv7zgHiwV9CYniNTXJQnU1Oy6Ub8TFmn4w2Dd/3oyNDOVt3GbS77y1qYTMox0XMNEOamLaCxztNh4t7GdS0P7namz2bOHbHQrjo1P1QLeY5NYBWpqdKSKs6to6I/rqCCWaJo/g4vIfDP4d63+twi zoxChXckgkYdvL0Xz4xxo34Uzut4Ln6X2MRnWwEWnHRkaDi3n52vCFo97XxzjG3m9ydH1QIKOilSvdSYBh+LNuo46xmKbplRcPONjoHKc3F8sVUfsDpaH+4Rb2mOPcNaRZSGMp9LgKVWEiQNnNOhm2VmWjl0zOnCEMA42oHcE8H+61hcIJhczYOOHdB/7SsEvznhMwq1ruvpGA1xv1r5ZdB9JGBWNG5oURWvv9dXcqltKKTt /dGoF5OitM7lzKned/k+hiIHxlLcBj5lna4bIoycujD7ZdjlTC9h+J8gsKDoRiZH4/taTdZllzTXCEaEYc1IF/1mrCxTRm+fnPBsoY/KBbMGop+ljeArV1x6be8M3T+xRfXKB9qJUthfEGTAL1DMI+Jdycstb82lTm+3KHcbL3fcrrv/9o4HEhLof1oLNe0mCeigx6y1Gsanri4jH1JXcCT3vJ0ZLGfNAAF9k3DeKZyT9/AQ BycwtpYIRIuYPUyH3AK1nmWV9m4/97dZL/g+7fTcKGtdbB6VY08jPrqw/g6dE8VaMwpTZjHH772v0EX2/aCnq+tYd+qhqWPnR+yVjFBPJ7eqgdcCPTYfFAJYcbIXpeDxcCAAkfk+aXtWWA+aGxCuOtC+PNxhbxzDi0D/gInf2Q4qeBldMhDXF+O4S6HuUByFAtTNpSjCwXWkdWWdxeeM/ZmXvYOprqDGQ6iJzvoUv9XTXQiU JZiqjkuR6RzAYhnOtJxUYDW4qAhzULK+0Gsrw5hel8KYByx2t6snqlmzC/HH/vsNjhg912Da0UFrl0Osf4+CkA3plj3M2iDpIFXsDwKg3DTWiAEv2Nu733SAWYFFGLYsn3o1er+6MlTMXvehcQPSLWrs16ivddXkOGnGRIRynlaIxdwX2ykL8P2zh6CIiWGrRzVlrtq5G88VhNs98aBx4VBaI5Pa78LlDWg2DPWzLVrsiUa5 MIpA/JwFzxbQIzA6DcDU8izF07EInwN4+YN1da3VQh8TIeUVoQdLwlgkHmFQHIpES0KE4b70XFcOSfRtRcfQUzjFfc2iioK//dldlmEQVOJ/3CnjdLBJ/hEIVC3MjRM+HllMu+oCB7s4gqK/jz+bkKxgevqBPRpZbWTepr0Cnv8xXljiJDgL6dtiqPZ0cBRAECF6HDAXXk4ZSTmugqEGYXtF0Y6g20M4ZwYyfTCtpFV2BJuz F7pC5iOZeNsfd6IIrC//k7IORFpFM70NXhkpBA9rbrE+2Mm5Ol0+TUVIXONTyxQYH5qlbBrGeAnl+k0mJXhn7QI0vm/mleJPecwiMiTKL7TbLyrYu7vZndrcHhzzyey8PEXLofL/txLaCoafsY5IdJHdhXSxrYZQEuUaM3xbAP6/FVZMAYAO3t1tbkFQ67y7L4WB9jiYXnMAQc64QS99HmfObz6aTy7BQSCOZ9UEXsQmtDtN thlD22J4WOxAqtsDiueyze5JAx9yY75+indNzj2cbLCxiZms4da5UtQV0JPywb8MDRExAFWfnfFSloG4Mf1HXX9IBEck2QuAFBWoohFnS4rkPhOgM/M+90mT4mkTT8X7KrTxziuZuclvrrmvTgwZUG6c4GIu53L81v5Pc+nb9dzqpxuUgrorM1YwsnZS3S4frYU1xeqTfW/a9tvVKe5HxQhl3OAEcag0NWS4/C+0JLEvsbGL 9ury49afrRi6paQXFmWD7WMO3yT8zmS7xUMyfPqrd4X04VyyFtWUBBvLRK2766Igmm6TQQZLTUgr4Qj7DHl13gpOYNcqsTjdoiESUkoW9rRmQXrQeWAsK0c1bZSwdx3l8yXelIa07QcurUpukCLUeAtTsXB5h8gLWAZn4n8hKft/W4x6dnKDyn56XqjTWgFWP5NNPys6okqOc1ZDTcqUgvnshi7iygPazy+aQxkmU5fT8Hn3 LCRcaS6HKoJ0zvGVQ6aAlOrj10jkY71oomeTYD7Ic+wk81sqLF7IsnjiNaVqSAg6IaM8XfGI4qczxhMNZv1VlSAjh0laPVQD2qW+ebu4XbGb5xhugXO2pAxYMH4/KEgCXNIP3Y0Kq1Q7IkpgSYFtEwAzkaVYWfeEqH8JSepluW+LBBi8M6+lDb5OrXX5CUXEEvXIVI5iMfp9zRWE31DFB0MHbNqnFNojkq1qOWVlCadjEVtn 953V8rywCc9/8ARos7LjcMby84FzKdMrmcqPo8gvKEoqkIYVCMQ11Akr9MDcKsGr4DhGDuw1kdftXbelRwccfzrWMJp+TWtJFWgi0bmOaRRdHpmKzg6bswjOUknlBHuACCru+49Iu3V1Y13oAgbfpVGbP/lhipYyDY28ChDG89jwHRjgHOZj5mf4Dsp4t72ZiubhtoTCm0DJDiKpvO+WL+KAjWhcAV7At1q3VxC4MiFMB/+b FI/evaARC8tYRauD5BJ+F61OvH80PL8mphBm2lXWwXNQJ5vSQrFDq+Iyk1EdKtQW4b7+Y9wxViJmA+OyZKFZKGJTewRsrtnao4+cE0jx7FyrNTalu/pzgiQ3R0rJ+2Cgyy6ERO/2Rr6YQvhjgMMoG9OlEqJhaUkYjji0OOqYz78gqL6KK2R6wyNa9on4mVIgBHQoQTIZw1xyjb127ui1UKeiMU/LjCuTa4Jpuk4h2HGgYcYW FIjE7LFXeRxmxMuyLn5gUpCfql5oBNbREg8EV3Tz6J1Nc2BIyqPJW4L3sWa9JFbMbbjTj+Uj4m5zFeJHi3AlHkwR42R/D+Ozhe6pIm4HKgow1BBh9dyxjmEhRQyAcj/Bx40rCNXMZBrHTQmjRScihab9Sqy2xqm0dQzTlWJ3oKl5nztekCUty8AjlvRjyA9ANz6jkYfLSdLuDIYfXly6eRegW4r+36sCDSllCzk671mHioLh Javpm14UEtvcazK7v+edusbFENC0PhMBh5cd2WyjeSu74vUIOtG2Rp//uNDNDijQ8RT5cbMtvdG5Ex+729cM+lnjKcAsiH37T4PUshQQoeTmFlpGN36cJCER6zKw1jry6R7FIWhGMqyxPC+nMIXJBGjauN8bSDSblQXI3oZhvC7EeNHlmjrJGxTiAZnYeaQvXrMRZkf8ZCRJKO02prQKgYPRUYR3n4cNmV3QGADyEPzFgUWi Fi4NTWqauUhYF1gc38WDXRpp/s53LKzaJZ1a1SrW8O7+r7w69x95GEcacIbyY8vzuEgilYMYfoKcStz/YhaYsjLm6fW6m0bBdQLM0z6ILopQmKEbRqzPlBNmNExP0Wdgcrym0dVzIcLWT0x708oX0NAi6vegAbur9fsvUpzPugOOWXraST2PcdtI15TNCHx1M+fLpzO9YFLQxTGethmISwne1b1yfBLXxeUST1+rZPoV31gH ePVyWGl9FC2o28y8sou9h5jT+Z4no+GzsXKKJUYU+unNAaL8JCVRFhk+Bnq3lOzuVQiZwJLGF3IPeW6IyxsSzQRR1uwkTTFL1V/hlseqNi7vBVFDM2BGNMj7Mf+xfRfncBEKJEb9qGOIXMBNwlBzbbQtohs3ohvF7QuUxAjNXXe2CiYTLJhji+Pjhdrw2cvwwGtw7bbcskulwnqsMqSC9uuTAo6KV4rHzd4XodcDb/MVCy3y ESei6FBitoZnS2W727r0BMrTMBdgwhCm4xQLzsmtjsny+f5sBlcgdQLhGf98826lt7KjSH7FZ+guQszZpKBugVQOOQRMdrxJXQ4NHU8/JTSH8JZnLBDUWi2wQRbNKP+EFQni1+aGFCfX8IC+tPgb1j7eh7peZL0b7KQfnwnPcuA+6kVlyl+ef7EjaQY5+27kfG6Fks6zSScyAI2R6hGsD6P6VJ2HfthYq6AWYNKeiBDjDK9e uuvReTec2pBk4bpVvOqBHtp0C7B8CoFYOjW9DyLMG5QZHxvRjNwK/V7RS7w4MOxkSk85aR1Kj3fN5szwkTrNGm47CTJPRyQWM56XayyRsc6mhHBazK4ll9W6wVFQONY7vSyRSSpsVwyuJHaHG0/xMX1EAGkR77h0y+AqGYwXminAIT2UtVr7qLA51T0pAOhUnAgQPtylqELGoqwy3qd5v1QAUqnDgJjGwGYnAMuAEPayLeGw Ey/8Pgj5tpBX7b1z4DJ2qhvevkN2GUgrxE0u9+xmjH8xZ1xdRdOrpPcEFcl2/Y2iWc1WgzEjWJ4RjMizOLYB7ArNnUHe45VgVTpSWdPqg7EBkiPPz8nbT9wKvc1T5x76C3Vozpw0x25q74synw7Vb3A98NkVCPwOAgceZ4dYV6PsyKDh/olJqvwy3nI+pXOO+MkiCEN4tQU43EWy1p6P6I+RQw5B763FzIgyzuVTSESigKSS PGqLq2AtWVSMY+sXMXcazGbykCuFF9SAuNFik1K7KpyMOi1hhDky9R6gF6isSJMy+9+FvyNcY0q5mrnIY7wXplGOzXLPTnVTGoV+Pbmp777G1wcf7QaJqfVBsnWZ/pPOMDBhHR9VLDJZlIsOfCklPJebSd5mSRLSXR678aZphVy6g9p6JLD66feUre4CTnplEkp7J8HpBrM2hhrbT834FR3z5hudIXfI/H/zmWpJl023z6ri PrMIacni6I6XrcfkUil3yorZTQ+ahzpMR+9zOef7/zbwzyi3IO2DMnRz7Z+A/42NlcnaEnvjtKO2cbOkm3MibXkSQa+Vf28bY/H+3ypEmkdYcPtjHENMK1snRTtp2Eb0xZDHZfwEpGR0ZTRCXT/lJHiabfKosp7/SExDa4R8nO4/Jyh9O9xGXGR2KBXblqSj5/KPdHJ+lDARUjgvKyJ36362uWz/fr7pk6wbSXMOR2XAyz70 n9ri+8CQBMbc/09FTyX5TtmGsgeJ07JFNnBQasYferdVgE5uetkIdLcWk4/ek8DUblvZICvpfQ2xMBQsPtp6lkJ3xqSbvjFDdUvfytKatBzrNNnjIWTKCTBu48XhZDYiDxRobt5rrn2yY6xrNwIjsTzZB7fh9T9wkBENeShglhbsmoWk2t/48Mj8eVkZAA2aPFbbTqwPTovV5BEoEwBKdyOvDVjmhls7MaFd+X4BqXoxHtaW LBG6UA+aCIqOvbwmv/uAyXseQmgG8j1j1m7ATe02vcQcsIBJ0jaI7EB8IWY0L7fmaztLkcPU8J0ZehurkQlvwl9BMpzwCQyZDyLzCSAZ7uqLXL/ZUqx0bv5OmIKlWmRVb+c5AvY/yv2M4RiDOQ31UdPrmoSYwoSo8VSv39C0CinltB8yYi9f4bfv6geC+uOf1icINYqERylPzW5Nh0OPB4bGA4lykqmDw9kwGBeF0nlE12eK fQTD0/CdmWiUpOV4Q3n8qT9fwNvGU5mYPqBfiKtK62m6vJgsHdEB/p+kaGAbFAkOWTnuFkHAF9OG9CjG1hdySf5kcJ47UjEHn6DMEjQyJulkTqfz2cxSgs5VCBr/CZ59A85aJvfhhk9RaWpZnitrtgeDlXyHvj+ibmazyXxnnWc+E/XodBs5qhKMHWxppHWuvGs1ChcaLUNIQXoTsQgT1jACnp2QSio/JHXq2mD48jfIpXCl 3O/tR6j0+hKkIBezHFpV40OOJwugJ1BWhNiNYie/Llv/BEFV5NorFXmCEb4x/DKv/kXlYHRCM55S05dsY4ug49IEPL9OXPERYo4uA4KATPzmPrQNKk++wXb5Vp2fKFXSz3wdehp2nqRUEpNAt/jfvLvFD4G2WugqGvQXRQw3ghmOfeZvE+OXXcWmrbZ9w5wJ7Ciw6Nq+DqfZg8FMi6BRo3FvVyuTVUotuRlN2t3k5dhg+ZZ4 9qU55dUx9ibOVQnCA9L3mvfGPVNAY9MSzaAC/HB0emGDjCUI2fhaolDl2QsMDVbOEHZdNDjdTG5R+CUaRFUPQEd6m6vhQYWLMLz1sPxSHjX5IMtrBJqp1YAt8k759d/dBFRqWxr1oiOs2QXhpZjje8l5KrS8ghs9tX33ab9uQQ71OGPiCXgtRvbkacO1E+G/8vNUnLyf3PkW1vVZN4lNq6+XNZobjUVclhJbtVQJCuE9nMZk JTqo8eFQNE3lOh6FdXaxZg9fyPQm0USCliimYwZmVq6PlcqW3bmFiZKZft7n5sRvqYbsMfOjxUUMJyftDonX5Dvx04DeVTfY5ljIrAMD2sTpSK5Zjhvy2FbD6rEr+mUSFtL+kWR+Po0YOzxNZ+tcuvi5+YWr7R6LFGsHq7l2dN6QtqWtau/zEm+/5wO/FKYiuYXmtcfzfYdE1hx1cZg1fg0RinG456ncDMGMGG2gvToGh9K4 3uKO+CZI8TkZBEzIUNSQ4uDOsS/koZHswpymactOsXAqlVwv5CBRlyRMLLJuJ3+iFlVRZBl3ZHif6nOF4TRyJX7zXJ1nHqvDCpV/ZY0KE/dN2mbybCL0XwosIAsXOHH8qPQk/rnPYxavT0Z/n4SYV48QCHDJm7Q2x3V35itY+J4OT27jgUsV2Mzkbao3XlTquyT/tHdwa7u86qezJ/6+RwHTvvzlPp8gcyQMi+9cWAg+qYyC IlV8nPTzyz3NDklw1yqR/1yulYmhoKd8pwHQFVAY36jvlfpPbXMmzFv8+UNv5sEa1jqFAM1KJi/Mm8tYVeTeN3nAwsDHsH2MrtAwTKDwsH0+Z/Rb+pNTbnPcsQ78JtRrma/fC065M9Zvl8G2Uj/CFGThiWmdLKsGXCmZ3O227M8ttF0nrZXdXhdQg9C08S7kDpNPQWdJf9YU6DUAEbzI9yRiNNHGRvoFHJxeKVkUrc6ltO2Y SGroSICTOImULezwRSAt2IedGQ4504iM0yRGuYNw/2xI10y8jDv+/UaCW3xWf2Fvbc3kIYcg/3XsVc/8P3FZaa5BX7h1t/X8iAdc2FO8wQ1DClLxA7TamAPclCdLLc4Fe/vyZsVPb5XPQ1pJCtslSk4AmxClsDppvGWa7kZ/wm7rdmixc02KmD0hFiq5EVRRxFdUWza4fpq8SZZpxrF6y0nvLQHoc3wu9b0OtNoGrxSUXpJj ScyZvrXgFgLNoaaqQW3DscZK+t5vsEEW1zj6c3FykbQ1AyFwDaxE88cz0+C7rJhbVAX6OlhGRIFzsO10KLer7Gq0wKWRtd3fOXmvTAWImPqDV1rzxfjY8/OQGRRcBM4B+9t+WwZ2Phv+1mbZ4ROQHBqDr2fGRW5Ycram6sLP4ogHD6oVIYqzihUpXTk+txx0Z9nw9xZDRDnDFItQt0wuRfcTh+N640FqkdrLRRSV4bo05rWP bNp3NmjBvspv62Zou7pviGq7bCWq+GYKA1XgUDinI7o0iuLNQbGXJJBBmRvmEXjAumh/d0MIfDIYkkNceJyPHBlVlchpZsQJfEGi30AOvYwFd22IMLhz/wtMv9qsIq7ha8xtrIBvNHfpB8ZdzIHp8XL7gPrsyGZQ7+RW33YFILOq5QxdCpdRbsYZof6SWLJCRV2bdTXVzdIhFzsspKRnQuvr/LuKG3t5wOiMcPhpaHtQC6in lbvYbckr5iO59Q/YuNJrcN4u4MKnGZVAj4QvEh2iK5Bb+cmIFy4RYB8QF7Cpr/MXCw6nao+49w8ddwLuPwqFhHCEUuLyr/9B15Oi6KGz/Q9ONPTW9qzKYHzrqJfa5OLAuB4AxgxhNEx7vQ2K1tN8eGdZLfhSy3dqUflEAyJdLdACa1Y0FDAzWD2ux/TTXwulXYISzDh1Fms8V8sxGdcxOu0+qxh/d68MvfqadZyrkkEy7M60 2BoRWj2lEjCtWscj4fYaYFVF4EtyrHAHzJks0m7bJKiUfYA3oHDjT6sEaOnqgbwXiknQ1yKbotO6o87Rg+f6KgEYZjty3Edn1XksA/QdcdXpmb7tTcFqUupbEzO7QxhJU/d89iHjhCUfWHuJ7yo+ahgo+69GO3lYDl19NWyg/h2Ve+CsjCa3Yar93GzXCVt1yYgWGWL7PBawCDUCJ561wBkPRN0pLz6OPthUwq//W1+f+ssU LDA1VZ1IXuoTxZMbeNCtEj1+bY1vw6UtJdVFESy3tX9gL5hx+KtuiUdxZ+o7TuNTuA71WIyvWNQZCkyZC3GvMexWhDh0FC02VClRNLLdVDaj49Zanka21z4AKCPSGOXg+Sr3gLCEaCU0XpdEGH7KNElZKMJM+un/UuLJ6FaDeX8cw2gvLaCu05eOhxUTZSMDcJUEsDEIEbVZMK7mG37fA6wisR5SKOjvw46Jr1GXT9OoNMqD FdHKHN8zmtYKctGok5MR28Qo/24Z2oV0lUaOF/Ro/sx70LLXeiHoH5fRfGbg5ZY11gS1kLdN9Q3IqEb4gkFdN927wHe6UOAj5/1xy3QUB73p7XSsdrSfgKOqN3sVis/5Y5Ofn6Wka9VWAQGlehsz3+yIyAa7chPMBh19DMLS5XQoWsLoTKEDbkFXVdoAdOrtrubyNE7s3oAxSd7bOL13Cco2a9fHMt4xMGbKPTW4kf8h/1x8 ShZGI9F0nDIgUPZz6o+yliyu7zm+0LMjXrEnEwSt5gjfcUVvUzVewjj7qUQRX2Nbb+iyhQWJvxB3fGmGBBYail90ulrebnLehDYm6WMuaQculYQ3mBfjPmFz6XU+aKlBtxUSt5zR6PH4a4XJcC7KB6XjriUGm7tV/wikWMCYoRdSm8wrAV9rKSjdDh1jItEhy9c+oIX6mRb+DGQkOxNHe4XCj/nCoYhe9V7rJXb6KGZywev2 4CLZodvZ6nIqTI5Gw6DXZSASdrsxKvn36A/NCDkTcgEt5I/4OeHXemhJbYx3CKkvr6Vg3NYYZosb5kURFqmh8lqeMRQ8hkewSJtb3KyOfSd4XiKaJIjMvpCZnSZhWpP6PId+i78Z8o+5JH6csBst+Nf5ADTIniZFIBYxpnIFjAMAAAAAGpAQznvovpQAAc6NAYCQA394ABCxxGf7AgAAAAAEWVo= ``` Decoded float64 bytes SHA256: `a09b35a1661456a42fa27935c4cb752249522d4c35984a51b41b71455828d3a0`; 24,576 bytes. ```diffs-b64 eNocXHlcjN8XfkMkS2VJEQ2SJJQlfYkGIdlGlEo0URQhLRSpkbSvKlqZUlpQSSqJhhalYqKoJEOpJEyphPB7zu+v+czM+7733nOe85xz7jn3ZTgSW376ZOsxS2fs89FsK2Z47ZfzjkYUM2y9xFD9fPwelNig/rWYYeV+/tlXUMxoWu0tkKjTY3iRX/+7zNdjCpI2hpm+1WMEeufcU8KLGY6VO5NRrseISob5PMH/Go2VT6xr i5mxiU8/znmF//0X9s8SFDMWoc7jVrwuZtS+DPvGFeoxLO5a84SSYmaESqv8W1ExIztusknH02LGYbWm/JgnegyjWP0r/wLmtSDSxikN87xpFzs8t5gpiF92YW69HlPq7XPI6DOe43jk4p28YqZCtMbF5UUxIx6c7iONT42FMX/2ivSYtkmKv/JbcX/08/qVsXjeEvXsgTd6jOvp7EuP3+kxUte8Vn+rwvrctrY53S9mhKs6 Xvv9KGaqDcefPAb5CCJzjA8kYbzR1y2V3usxbBv3wPIbGLdz59t9DXrMq8ao+rEdeP7wilXXIUfRfcXPx4v0GIfvR84qtegxfttmztncqccM7mtKycK8HPYvCfzQgPXazX72shbXjVj4d/0DPUb2nejO3nqMZ31ZFIPnV1heWHOvFNctOXHGE/pibg1e6k4pZvjz82flPMZ1aV3HGrFu2enhkZ9K9BiDFYmB+58UM9ZjjMyL mouZ7CbXNI1yXF90aHYSrnd4eSEu744eM9S7WOcS5F/Q/XR1XV0xo7MyqHz2t2ImPrcgOrazmJFWnDfh8ftiJjr8vlUA/vf7HX924DXmt0vFfD7GYbwE8VOhJ3aoje/XN1i3U+ylEMjHsOd4uUoHcPEodHvBrWIm7caaX8IKPUblZTbbDPenFZ4erl4N+UV3bu2g5yzYk3PTG3I0S595slSPUbs2re27sJhhLlXPOOWMeatu CJ+O77LDPpwfCf04PAlznoTn8Xc8T3YCPsR3fDaOwnp7ue+XPQd++BfX/46uwHoyYjNXAn/MxI8LbidD7sOva13H+K4j003GAw+DC/brjQCes7/N2RaLcR38+hmPe5hP/iLpDxewzqM55osHMM9pcgrLoAf2Td+JRnFY78Hly6wEekyn3KJ7z6Fviw2Pt7wDrhoaSjRqgUex0pKe24WQ71udKMl3xYzBpDdqCx/rMcK5yk1d 0IfDFNGFw8/wvFMVZkwycJapMfFLSzGjMDR16XzgUXDUI9IqEfhxPveOVw37Wz90+zfkxfC27elfApw6Xi/wg7wdRrzROQCcMNP0rsmFYN15pZO4BcCviU9CeBg+oxSNNgKXftZvJu+Gnl15+zf44tNPq6rpOPTL3vHf5qewS4ebZ35+gtyY/ccbDjjpMba5uvft3xYzumVL3OZC/2F/DEf3w/4ZradNdVh/9IPSfHnYOZN+ 7le8D+TX32LVC3w4yIVe+f5Uj1EoOFTdAPsWMosLlHNwn61NR8AjXD/x7Ng3kEtR3LORTZ/w/Mf+EqGYh5/p6ptnP+oxyZP/ckMx3/gPzZKZ0I9GPWfyBvxv0BU4qN0IeQy7t9kWOI4+YTJ5BPQq+PLTbgF4wGJSrk4v7Evk4CtWBm8JqkOsV6fpMd1nh3b14z5mqSOXZ4P/I1nxig/1GO4RgzEbCE+vFowX5mFejVVvq/0h H68PMrpNwM2HY7Nmv4K+is6V37iL9ax7pH0a65IdFn9qRBnsY5LN52LovfrGquFRkKeA7ezvzy9mTBc1hn5p0mP0M8zjhFiHsF11ZBHwlOau1uT6FeN43t6RBPmxci7cqM0oZpSqk+YXvyf5pHMjgI+C8PwbI3Cf2NK+ZjfsnBcz0i+0Bnr90xM6Fvwo2GSt6Q9729VwbfS4Zjxv9wmTIMyLVeSQeR56F8xnWjqz8N3re+jb Mj3G/PGbgSLwoljhwiLLZ7g/c6atMeTGq3NN0gaucv2eKjcC/16dR9/6Qv88JbnfW2FP3OUjTmgBb4LfW1U+pENO/zxWqoJ/+FP2jSjHOoNilzxMIDuLibhe2wgcP/NYaJiF390mdIghFyZ41ZWii5jfCsN/lyNgp7a7m36CT5mRx0L2wS9wRi5qq8b4OkfV1/S9xP1P/GJrovWYoNqdJ1Xbixnu10dmRx4Cn3Y/D8WDFzne 1VNHFYMXUu1TT8LuxSF1OWqVeN4u5S8P4jHP7/c8OzLwHEHOgXPl9Puas3vAj+W7FuY8bgXeZVyvt0O+Uo/Wlb0FTgfHMZlPX4CPFnF9tMHD1W5zlFdBb6KgnEe+4GPTVX4/iN+FnbYTnaEP16/XNyUCjw1fDUsv4Hfu1gVZi7BuXlPld/lgPSbS1cniDPmx7O0zpkIPrDuKx3YDR7av3AoloTdV08hRrvAfIttl/B+wC816 zfTnsAfTdY5veuF3smeMNBqN9V2y0FySBf8g0jpZM+4G/GOop4cd5sk7XTL7F/QZ9iPB3hv6YsS5+94cw6dPMne2P9bbsmn+KuhdeEh55u7neP7SbAk+8K0Z+7ulCnbBy9BV/psEO1J9k3wDdmtRvslyiHB35k/nG8ybI9Datwb6YcnLXerJhr1vtd6SAv4U7BluxYP9chybrKpgJw02wXvPYT4697/nysDvM59mj7oK/lJa eeo8G3xhO9tqniT5F+1ZS5RwnSBm1MBt+CV9ts20T+BFTkjD8bfgWQ0PB7tZLzF+DTe+Ev7RNC+o6DnwIPpyet4Z2GeFV8f1W8CXUD7Y5RLmIRso8T61jPxey7MkPvnJbU05kBPTrWg0uQbyUnfzngV7GEyR9PWB/qXUm/SnYRxT/duMPPQq/h22xAh69ht42/Mcdp1dtnnqMdiBtbVTpMIH4CpjRPY/sqNrpd7h8Ith0aft Bx4Rv8jaMPdgFwtCM/vBWzpHJWqOYX08n3kLNqTqMfZx5x+mQB+2aQuTHCF3Xv4nlYpI6C12p0ERvnN65hmfg18SzbBd+a0Y8reUWGUP/bsGlc76gfmJVE+amMBeBKHWB+9CDsyEhROOX8JnRczRylTELy/2jHAkPl33odMDcmDN8b+YBz3lniyJUsRzePcexPEuQ45hd3r/gJ+lHjCzLSEHU69L8YawY/ZL4RrNBMjL3NBA hvxNjUPRkTLy1/meRkW4//yXjkTIu+nuu5J3YsyDmyUzB3bOU3XZX0u4sx2TMTwF81x57uNfyCV62r0cT/gz1z6Jie5Yd+TJowZ5sIf4X7uyRkLOzJTDg52Qv3Wn0pMIfA+7ZS4cA/2opW+OOkB4WOEmXQY8yYbu/PkV6wlLy1/9CfbPS/LsqoJexdab1P5ALuyiY9y/mO+gyzrbJ8Bxdq1nWTL8VJ3z/P3b4DcYi9GlOsFY h5bLJBvEg2zWjl0zYG8OU5cuWg07Z2mPDTWAvkzT6r1DgFM+702KEvQge0F59nL4E2b4f9u7PTGv5WkKIjyf13b2bw14hFnp1foS9iTVuPtQNvAp2PLAyhe8Jq6JkmsCD09Kd8lYDrkLCibZ7US8m31u/mkNyEsQ1fiSC/uLljVwqgI/hZ198q8QcQrbKrak4Tb+v3hjsB7rEepaSG4lXh279soT+O/o/godNviG693TdxzX 2TquTbeCPCou3NJ6B7/O9rSqK0d8xGV9W/EHdsnUy6YY4fmcM2Vvd8OvcfbFNfcinhEbbrIbAM9o8kd75WLenAM3Uv4AJ5FfXZocgHcxb979j+Axlvy9G5OIb4tjjP7CDwkez3HpvI04v6govQj8xptmor0GvColLxkSDXuVym2veA4cCx4eWnqd7D33D0v2DPD0OeZPNV1/L3bbqOeQ89WH71pgT7ZzmnnJkLNt3jW7L5/g 5y/4H9OguNlQfD8Ycha9GNefnY917dFKlsf9zIcVPfcgR3b4DfXr4Hn2I3Ze3lU8/2eYTRrwwS5d7LIW/o5/TLv3D/j24tYjY9bBXrkvfZsjsI6CwFUrJwKHAi+Fr9MQV22ZZj8xhfzFqsmpVrCbsdOsItcirmgLPLL2PfhOFPHy5Vn4szTTqIlXMe/mMl6bHPiXN6a+bjHGFRiGvhoH+Q3GXvE0wzoa2kXGuyiea37Hf4B4 TzQ+wmoQ+uawWk2MoA/xhtphB6EnnfT77g2k797DCXsRv7lXh4wbCZxq7lx18wJ4XTbr8SNf+F/NKWHDpoP3ec1M4fXLGG/HkmlNmcDlFoP6MtzHy/x+cDTsfrFsicEf2JWUtmSELPIl9ppzbQugT/a3CLX/EE8H2bUcJHxK5R2RTce6Ogffvwyj+OrPiMN8P/BRqOTsYeBLzS2FT+ZhnmMb6rfyEJ/6HXtt7AE7t2BxbsyA PfH1ri38B3tc+s+glMF4vJFLRxXB7/AnJFw5jvkWHBAv+Qe8O9irz2DDTwgnuKkUAkcM+4iDrxWuvy/59xdwxHSdTj2HfEpNRd7lD+yd3xqT5YP7ma8Vx5mzwPf0pPF9OeDPvqJFu2icsazuTbhesNf21qtExNHzus/HQs6MrZKUCtbD/VbR2nkX/y/9ryz1/3Z5ZiflT9lP1TN+Qs6i9vLzfZBL5LfDdqtwn/1ivaqv4Btb FecXcfDvsvO1hu0GjuPtNKtPgfc1H4sSDGFHvK+K/oOwB6k1LvMmwH4MLjRofML/XDU11jCy8+nOI+8Ab0w1a5fAC3as9GxEAXDEKcrWkwL+cy12vnCh+Oxg2A9j8CsrdJ1dC/I1gbTn+zEPMO7czzrrMC5PU+1fTgxw4eaX8b2R5PHi3wmsx8FL/6gK4lXmy9Upi0PAFw41J2/CbnQKIq2ioH+/VuWF5liX4PQYrYvgZ4fl Lx4Oh57YY5NqL1BelPVxxx/IVRjiWrIe8xQdWem7Gnp1TzoeoIh18SKW9sshPsteY9KxEP6G86S92x1+zF14e5MQdtM58HrvEfANv1IvwBh22PTr+cxa6GO119QzE9vBtyW2e4SwH15lMk8B8xAZPM0zBB4qojevcIfcWZ/MPM+AZ9K8dirfoLyp29tNG/6AOW4jjIZ9sEe6TbGgOCJp7DZ38Jxm5Frnp3ieKKwrKQ/2INCx PGgNP8VrkDHLvY5x/Hoy3CpxX/OdHVsQR5ieXnillPLZjSb/csGngtKMkYfBGwpKnjbjcJ3fn9sPgmG3YvmmDMqLBre2D3cEXnT3RCf8gn9x/eSY8w/2yLKcuuwu4na2cnpFE3hEmHZaieLxh6rGzbFdeozzlt8tkz8WMze3bu+2x32DYWdnnAUOBNq+TaY3Mb9nDr+2XwHuNp2R/gI/wjQU9fCAf+7oH6N1oAdx+5OhQyUk j3ytk/A//NyPuYP3gZeq7XNGUpyyLcX4I+bfue2UWh/4uShAXzCL8tD+Fz/awB8V1skv+V3QywKd/QLoLTnnVW09cML9st7iAPyA2DLl7RngyPZ0nncb/ufWZ6kYY3ym/UqZO9mtv+rC28Ad92RB1hjcx0pzu+oPv8vuKbldAXywns7IYIAzrsb1bfrAOfvIo1oLXB//+qtRchfx+g+ZZ8AvP7A56yHlR2kCl42YN/f7YYWz iHdEHzpiBVh3mnrv5Y3QK2fsvqRerFecfuCxF+xQUNES1Ylx2ImKqdeRp8rKVNyeAjuUdb0wShv4qFhbOZINuTrkFpmOBk+LO165s2GXpX+eyOhBDmmz2hfHwF651tWvVcBPbGbfwWOwF/4Yf6E31iHOFDkd/AD7PJU6zZPylVFaccHX8P84lwRH4JpZ513bT3nex+74/fBfzPvr+S8obrlYmjK3hOxmvcEurEsst8UlF7jl hgYat94BD/hfahrE9ayU34WTwLtcUdzrs+B/tbnaj/qAZ4M9g2erIc80Y32tzbB3Hb3s0ikUn4VodTsAP9kjtzwpywXONOQ+W8MPCx1exM7AfUG3D289AdypbA243ACci2IsW2cDd7q8md6b4X+EF4P4xylP9Kit96G4Tm6G1xnocUvY8dcq9PvVn3YXkzBPCf19oYiDmNWh4j/nwafubpNHw681fCkcew525Dzlss2e95Sn zNHPR7zDMf94iQf7YPUY12tjPLXG94lJkHN8uNa6u/CvrncXNNuSXFxeGnjA34h8G5WtoBc18zNLoqrw+9uO3/9B3pr6G6T/4D61FdNtn2E97K1VjD7sP3qjbs9X2Lfsk8ueMxD/a86xfVAF+TCsEsf/bsJ+lbzHp0Zh/ZV/Tw5inmN547elkB1p6CXogs80tb+Ii4F/JupqdCvyQL+Xx+7ehFwtbLd3XwdeBK1/206Bh3gv J2b6Q54Wg6PepIKHbNc5HO/B+vgHL/t4foafkx6e00fzuxb0wgafpu0lx0Mgt6LFNtuG4J/qYmdpHYc/rPit5eYKXhR4d4WFgi8Y0f6ROxyh5x1fdU4hn7awWvJrHeJjdrxxciLimrr5SbP2UlzSMFtSH3zL6MycN9wEeU7/WzM52IOgwfRJFfDicDOoXRHr2bVhco0+5C9ICPeyA3+lnXecGYp1cO0uWNbAbi1qDr58gO8G d5My94F3eZl9jl+uwc6S+mc4wE44KTe7hiP/4ter97zCPDTV06oSYV9+meu2GCBeYu7Jv+bDT3KP76rIasP3uUf0JNP0GPXJti/nYdy+usOdMvg0XWmyJRvrV5i64Osf2ImmYvBKeVpHruXQ/f/vG9k4/IZd8UozqpbFwC6/vVIefk+PqUsv3q4KvuTktFs8oXhBZ9Xub7jPYIR/Vzf0LN52M/8G9MVp3LZkD+TVabk0wxB6 qDv1YUcg8mj+0rsBJQUUj1jYj6R9qakJT5Jp36i4s14GcUeB8S6VxxQnFT6a+gm4Y+vbudiAxxqUPrY9pv3M6O4FhpBT6d17R6ugr7ARKumliI94dWcTjOBnWCM9X4+BXoZG2Pe/b8e8ZPbmhOM+A7sLB0YCzyMN9nOmwI/xnFLm6cE/q6XdHYyi+HLhpDWjgWe/nbp2JS1Yd9M2/TXAlebzhed+Qk7R1r0249tJPs7MdPhH fumgkxn4g6Pzav5fytNark6wA+54r9sPn0Jel93zmi8NfhLqSG+LgH3Iln30WY51MQ668tqQh+hy2r0XFIe/s+11pHixOGQ48ZqgmnN4BvyGa82bOrJ7v6ZPm+bDzrmbtT+og2/0p0tXrcX1vHTP+KYA4Dpu4az5tH9nzVnItsP9an8M76bAfkcoVmd0I8/bK8xNpvUU9l2+ARxJ/ZKeLQU5CCf+UHpH+zSTIz4uwHw53ABT KcyzIf9boRbw19cyaYsJ1uM6qSFmPvAm2DM3gfJDwT3dhj2YN2NztMQa8nYdf1P5M3BY4D1PNQx2H5Z4trQX8rDPy3NdDJ5nqUee3oF1qXUP1BzDenRWeUwbBV7gbjvRr4G4UadMscmW9lM2yq+4BX7sNFdyOY3r3ddt0v+FeXauStufAD4YDNl4jvYFRCoqFbHw334tDww/0roc219uRP7JFNrFiQ8gDs28MlYLz2PGPfNa VUr7SOpDCZSHf/4Ydwt5KKfa/d4cxGOabToz/lId4orVb3P4Be7KzRJD4B9G+m7THOQnYrMYV034K7aSy7ln8MOyXca5AuRfrHH3x/9DnMGoJ97XPkPxSu7RE5ifrLHXYCHxhceqp9eQH+hoezjOx3p4iqYtH2k/XkOj7THWZSCtNFAGO/FOnuI6FXYi0PgWXQt5akbNb42CvDisqI/P4X8Y+f0b9LEOUae37ynileybG+SM oa+1nTe48EfC88N5zcDfLv7N2O2wX5Z1y9g+xIe6fXWS4cCzWNGk3hRyYk7/iAy3RTx3ev3rjZTf9b8d/Al/2iZ5VoL2L1X8otdcgF9QSL/nPRy44otORMvCTluGHuUtFOsxMy4vdcsB7oPMj2zIgLwG7x60fIb58iwUZWdCPwquB6WvYR2mlxz1oyjuPBETw6J9ukOuwdZekMvVjfO6aV+sP9FgKuSv0NJxtQl4FpZoR16h /dND9aGXIFfRSd6NYPBlw+ife31fkR4qfeaGQi++pZr7aL/EsrWR9jeto1t1d2G9folxe72AO/6mad4dtB9x4sZuw0vAmbfJoxOUXz/Pur40DP/nu3or4X5GZfGrBmuMt2b9twfgHUFw/hMb+Eup8eZ/WLBvh9RTsVzwmYJa5oL7xDMH1DMm0z7gzE1FhhngG7GN1lXCt+SWJT8Qpw/8YzLXggfZ7NAAZfKL8kqfPsOu21b0 9s3BennKzKYNwA1PZkXgCfhj7gGjvxsprtI/Wa8EPxvtciL1Ivmhu6f2BNVAL9d5U1sozz42xlkL43DU+kdSHOwuMx++G/KXaKj5D7zNHPArvgo5dMrGjfXCdZ0Ku8q2wq/wO76cv4N1yu7wrfa/o8fc3N4p34/rNfUNdE/CzriyTyqTgE8pPqe7jvye1IjLE/7Bz3AlBAbAr3Bb5Zq5sCf26vTrl2gf3NnOdCLtOzE7FddT XrIpSzD7Db7/O1iSSHnV/q4Z0eBh19DCU6rAE+NUfqsY9lLQldzpCb0W1LiXrwHOG8Jr9U++o3ij52Qe4hbnD2tNR+F/zqGBiY8x37QXC2qDMQ7rq3bycORPnBdzLtwF7/L7dPtryb9cifUdorwk+6c48Aa+nw23mcqnfHBc1z3Ey4IHVlls4EB0tP5XXjntm31K+AQcRe9XuDuO4iKnwsU/KA+RnnkyB89ZMRh4PxVxu6zj gY8ayGOyJwTXLMT47Dty24dgT6K3LzdPR/zKSg49/wDrK6pvicuiupZgYVQT7RMEr30YAHyIp/446op4Rvj6nvQM8puH3Ff8xTgss0Mqc2h/xq40WBnxTfZxI4uHlF8qzI/a1QG+8Ao8wgaPjxhkVn94i3hjeUbWe8xHpaXn+grgWodTwDKBvJz9D05xpv3LGxvNKqEf0Y4Febwi2IfniTUPkK8JL2XY5GN9YsXI9qOw35ut wriJ8GuyY9T36MCPsUce67aEHfGWmQ7fBv5g3RkcexPxq+7Ylj+ReJ5riElhPeYVZn32dy3yEd6jYyoOkGeXrNvTheCPNGH8NR5w4BqQ75pM9jfrwNUrsFPByJ2T+LegVxnLfS74X6EhwscDvMl+O0zj6XXYy7kPanHB0FPet3vZtA9V02gUCrmpHaq9TnzmOqH85iDw7sD7HJ5A+xKx943ewF9IX1pZW4U4uWFrtFMIxXvn i5+XYRyxzBfJetg3t9rsOcWTN6e/G7Ud8ZymSlmMD/QnshSpDkec4b1lv4cT5DhYEjVrPOIF2cT/JFJLqW54bpER4iB2SObs3ETYVWDdOYq3GN0j9duD9JhqG/7n4/jfoSSL3414X1Z/tJov+M10zoZBEfiN7RZ/MARyUXuucH0c9BKW8zF5MfKizleWRlLAnUpGG/MddsYSRtn6EH+VJkW/A84EY/wi6hNgl+fGDrtAfHkj 4cy/O1SHPurwyq+Y0U7R71sNfWV/N5wvpH3V8AeWppAX9+azLlP4Wde8f/ersN62EfP8gsEzaq+upOjCH1dIR3SoYDyd58xbBfBIwYvN0gsQT+vOeHtAAFwWCdlz5gIX2aqrHm2lOLrSKXcecMOK/Bo0j/R0+E5WP3DN7v1p/4b2zdaoNTbDj4sKojosMP/ui88v/sU8hL2+UjK0b7bhqKMy5NVw98zPIfgdwZSe7eWIL1lK 3eG6eA53e6ocn+L5Re7rM+BHG3I3Kn8m/5uhplNyEs9PT8jIoH3YF//5+QAvk75lqY6gffvJubk9F7Fet2TDpfCPPA+zrevwqTLZJGMaxhfcLthwIhr2GLfNzxO8ENZyUr8F8uO4mCn/Jr/Oixrvqwi5Xf3lxYN8XRMq9tkjTgnTZh3lwY55P/Zr/OUhrzi9tKQWdscPCZZbQvn8rQKbI7Bb2fycy4vAo9zo3VUN+D7iacKY fuCuoaRM2R9xmrglqN4I/lp3yiN3Cdgbd5eWqTbxSM7zU7mltN/z8fYjPE+2/v74ZeAPgblOKAP8sxWHmx+BfxE/6OUY4/lMdtgKIfUd/JEekopCPj56dsBF6KdZYves7dB/2Ithi3fTfp6L4Hk2+F6Q+Oor4b3OTXFHI/J7blf9swrEbQqrx41g4zrBqRVV78HHChNPHYiC3YdlbvOi/ankoGWvX+O57MdXhwzBK9VaL5b4 tZBffrD3FOyX/SJXzoriigLlgUOYV6TPtfXHwC9hD+0HQkkO88T+w2BvBumXNmdj/d5Z7gseAC+Cda57FkG+0f+lyA4Adw7WnLvbMa4g917lyzTw1dMOd3/4H4eqDfNn4HnZGntnq1F+rBRazYOdSHXs/rYEcuTuff/yI+bNUX0eJwH88M18N/7EOgXqyY8+UB1YXs9yAPbCnj7wbzb4QfRJWfEj9Mi9sqt9HeJ71qKHLf3k BwxSQw1oX7LAQbIb6xkclrTRCuOrLDw5zQzyFPj8tRmViU/WbaVx6cBP3/PYuZTnj4vtmkr1zwOPU5No/zrCZLArD/p49sD4GuIF1muZmKOPaD9oR5su5Bk9PvJvLfQoe8v4zCyqH9g/d7GFPQR1GTXNJt75PBjyCc/rvPpsWivk5SqsPPwK69Sct6WH8uywnIbZ55Hnh9UdmZyB+RW4hF4IBu7tV5fPK0I8xT5e+G068jJm 3LGGhYhf+aFRIinMY8uZtLxM2LGw6Mkzg4fkh5ve2EPvBcPFXjLwv2q3VhuIgFPeQJLfT4ofTs02uQ2/3pl0f8VVfOcumf1DCXka70uHRCfx6/nF28ZgfrvWfVlxH/NmR+n7/YrEuMeN06/VQA43Ss+6Qx78oAShCfQTZmw+4zL8RiTnt64c8eS2Z0rptE8R+S8jDX4kbE1Jixx4L+37F1E27RtcKH9pSfWtPNfGPPCJn5bL /L5OPUb12abDNfBn3fMjJOtxvbg06rIbeICneXNCKPhU1m9kxmT4Sb8ss03nKR8evF9Vhnl0C/5pmSOOqrZ25NkgDk3zdSyuQBxdMC1jaTTVqy7eHb2S6iSPrq5xovpGx6ikRvC9n8nRg3cpzrPzXC4GntiSuU+OXYW+d/ZcG7iJeV7mdhTSPJeHaKVCL9nRFUfuAvdb1k7Xi4ediOsuX5oL3LHORZrLgmdEH9JCfCFX1+gu 6aUvgEP1s+MXUz9Msc8za/K3wwN8X9H+ZWVSRQzyKL85HgHLIF/xsaR7C4Hr6lHj3RrhR9mHZLjL4N+Y06W1UvS8cUsu9kOP3i5GIUPUN3FAcdQJPNdvo3C6LNbTNlH1YSnxp+am4mvgF9GwKdUVkLt47v7gD/guOLarYBvl+/9OswdgD5wvywePwA/wwwQSW8DzaY7aBiWwW3dvxf31VE/LGR+ZSOs+oGq4H/7NdPUxDyHl mUrmgQqQL/MwdDXjCf2vj4v8CP6oGG6hpkr7c18+D1hhvmHlOr/4FM/8nDJhHOYl6H2eoQI+YustjxdBTvzzjmNCaZ9gWur7qdC34PLwNifi9V1SF6dXUvzL3tcBe2cvb3DejjhZuD1sVjjsaOyR0y4ePZi3YKadA3DXrPL1/UP4W76K88gofNdcsXmoFPMTXPrDmxaFeRV3bzQBDhv2XAh4hHlOOvfszh7I95L15bgZPXj+ 8zsv2LAnYQezY3whvh+0X2cP+XRefvOyAzjldlhPjqL8KMebycNzc9cv6af6cmT5OnUJ2ifRGVjREgN8/zfQVkh1u13bP8pBb6ythsfNEY8yWy9qCan+MOW1+ATkar99pe9L+GOOzQaJKOCyuq9bQxHjOM9iq+8BD5py7ftZsFNuR3JRMsW7jks3NVNe9F/e7k1Ut/xWcTYcfkTnv+5z0YhneM0NWgGIT0ZMmDjHFDzGdKap F++F/LK+8ndR/Ca1N+4H7uNaFqkdoDoad73MB1M9pm7MV83ZsJ+wptdbSvA8oV1f1BDwzU4oClWAnjgztbcqU70/7Mh4szDcb7HEOwd6aWCt74qGvVgrfS9hU7700+P+3ELM44ZjtgX4wd5ZJcsReJS61HPjG64X6FpZXoB9MDl3NRPsoaefNQuaH+L6sKTpdlepHr5XOYvy3wDbotUeWK/CeFNZ6vdpmjXB1x1x7ptnSffA p4zLsuh/wJ8oXvT6GOKJbde0lm6H/GWZgPKD1EfmPnB/P/JWruxT1RCso+HY5p054M+GWiPDIx/BD34dJ3wQb4imTtDZTPvPf/suWND8T8ldnQ17ttg1aXg+5s1/82h6SoYec/Bt9IyuduoTq1WXf0z177tmW8EX7E/vOsLj9Rh1N4MlHMLt0+DdB5EPspa7G06k/jgHxTdJdtBTQ6DGWQHFbwdDHiG/lFLr4x+hdYaMr9QK hN8RO19zo3jrblCfO9ZXt+OIawfhR/2401j4rehtZVaLqR7ScaBv/hX8Lh34xh35Cy/g8JAu7TsoDV79Ts9r9fD4hHxVzTLkxl6KL1inro6g+lfDOmHecTw/TGxwEXF6tmKk2lTIj59a/M4R9mdgvGP9M/gH4Spl42TqQxrxaP9q+A2eaFHOtyTY0VGbXsrf7V9NVqqmevxjrVmeiHtdL5988pv2afdfUilNpbpZ5bIeD8j9 xoDFmmT4xW+qvjzYV3S06+Z22n+6KP3UH+tQuyrwv0a8tdJ42ADmIwz+LpeGeIM79m2gzFN8D93jlAW/KLRqFb/CfAT6IS55sEfu/k8GJzCvgoWbnyrBH3DWBBjKEo59ezKCEX+G9fn+vUx5yFG/2Ysg14LvFyb3Ut3q+dQ51ci/RXPfV0dC30Myo2wGu6nOElq6DOsWdc9OXA7/rmQ4OSwadiewtBonSoSeT5Un/IEcbYeO J7AxDvtKWM8X+B+HCuXd18hPzDzWlEJxp3DMpqXAaenoe2Xl1M+yQDvlHeI07tqcxdtp/ivWyi6C3Q67nLJKGbwh+pQ9dj38lD7nm+4p8JWDnvawcZQPMz9n3U6B/UoeOuWH/5N7HJ1kkY8Ijmcd14E/Zr28dH8yeFjoHiGxGjzKWnLCWQXzUzP+sXYH4g1+1P7PUfBbC13fypuDFzWnae86THU6Bd86Cdh/Z8ngZBPwyVK5 7fsuAd+8T7+krPFcruWBx98pz124PMgD8s3+VRh0F3xt8IU/+TPhp7WzcXQ61WNOrj0DvxTWljNxH8VfrlLWJpB3gVLxlXDwzVeds5pXxHpM34yQMcXgTVtu5MYJ32CXnwIv9ZDduExM64K9M7OvH13rDb09sNogEQB7yxlMYEEOBa+sVvwhu7b4b8IzqqcMtLWyfJCvFbZWbMPz2FWOY/nwE0WvlI95Yzz+8cx5v6AP1417 BkU0r6UPbG5Svtr70/w87MC2/6fPXsoDOs1PlwFPPIsG7zLgD4Ge2zys0y9W63Ec1bskFqTogF8HW5/LKuK5rjsHar/Bz7mXrhwVinxS7YZzfRbxY5bjTn3wjbh1ZI877XOYyTt9AN8a+CgGGMNenS/cVLKrpXpYN1MFfxwWaeu9HfPSvPZZshi8wuf2LvYEPzKZ72/Xwb+y2ocUr6ZAPyunHjoO/yBOXL+vjOokXve8FkG+ anGJihHvgD+Hkl8zoF95HV4U1U9Fba9mlWF8KYu9N6NIzkNabfUUX9hpCV7Cf1e8860fD3xItRmM+EK81nKq6xziWO6kLdtuwS+63quOcaJ9w9cT78Sk4rrSsaM/QD+D1x2/roc+WQMhYZPAb2pXR2dTHKpW+fyIPq7322QTOpH22R9nH35zuJjRcNu7Vgpy4t/xvsIqB1//MLhMfRd84Yov+xA3Ovs3OV5BPBP9N/D+9s/Q X2LKUD/VWbr/rqJ9co5x6ZTf5A+WLDnVjvhKJzd3hj/WoWn10pbqMjznCJNU+OMwj4aGKOivIEZr0g/KG9bFl228BT+62qfShvo1osaYX6J6795lP74Dz9kBgQ5qWKcg1VhjAtatueFusBntaz0yPq8N+xEv6783DfGgtu7EXYqwL4WIvmNnvuL3U+ZWxDs8UZNhAO1nSTPN2cB/2vjaSx1Yj6yT+Ygb+J+T5j25GPpmF+Qa lEJ/3NgNGsuAW7ZCqEEI6d1auHkf9Kg558nJQuq3mbnuYCLyDdluO5VH+GSP+idhDnvrm2jutfYr5su8PyEPPDaM81hN+NJcnekyk/rm1FJ1jOGvOVwVSX+qR39QHb4W/Mib/fzaROSJmbdnz7pI8c6l9eJi2IVILic2D/Zv4BfgvbGa+siau9ch3tPkjLyViu9F/rnXf4HHwjT9lw5AXjyG7fQxFuO8a56WRv3LuulzJcnf 6m378oP6H5KeTxJRPnnCIekg9ZX11x/ccBP3b2rP3Q397WrTkrWBfTiYhh49Qn0Ny3Z+Sqd4ftqsH9egx+w4pXfVxB+312yuozhi2265K5hPmo6bZDTwJfzgaJWDfImJZ3mvpz5C1cM1NZBXzsCSExzY21inthof8Lz+txy1YPgRv6KGLZLQG3tL5ZHwnGJmhtum8Rm/gDeHRyqjqf9bJZ2XAbsJKur25BLvp7oMjcd4/A2s 7734rJgSNZD+ivo4Hi485QW5/pc0W5Z43KIwrgDyF9v+fBSHcbNdp5dPQ3wtHB63Yy3WaTtu8w9z+G/NMx/fuUJ/deoXFxnADjTr0kyMMN94pS0THlMcf3SWXzXyIfdQ+bPfYC/R7zeevoh4kz1++Q1T4FLzYk0c9SkJFdeLr1RSfJHb4Q4c8Ffd/tWUSHm+kZHVIcivpcrgHNV/R+11PUH7ohrlXC03isOn7kwDfqXWesoH 4D5ez8XpY6j/+sAElzKqv0snKIqpb/z4kFx4Aey9rOi5FOyAFdaqoQHeTZOeZkv70kzQuo6/kLvm8qUWeYhDTXVv9DeT3+trG68FuQquem6/D3+hYPWhdA6dU3Cru6sFvEmtqd4RSvvai2aMN7lEuF2eG/GE+ojyVowKgH6K97s/oP6hfqHq9wTYy4SNXFuKE/84u8lRfaPp/bLl8Lf899GmsrQvEJK2qRzxB6+w7WE9cOv6 fFUV9ZMw784LU4Hr7JSotzHUl6ckc/YY4hA1jy0pqrA/fpOiYC7WxZNY6bsyRo9RmFOVkgO8dfaq1iUjfvK7/ayqG/LljPcdGwt74hs/Wa8Hvfkd/Npzj55Xufd7/C3ioXxnDaxDLOdUdB72Ybp9xPYW6F92f0mMSQ3lNYv6flCfyJk6tSrKB37Kz10EHIvqC04eAV4GFbdPvviS9iWCRhRRv1rk/vQCyFNW6XdZMnDjvHnN Omvqx7WP6t0KHAiG7fVaRHUkV/9DmpTHFglnboU+bfdtHalEvH72iMgYfpZfLrF7JuKVtLa3mVLgnTCOz/0ezEen7XbCHaprhO2oeHcGejX2zJtC+cGC+d5GwJXDJfXPtzCOtQJrzAva//yv6YpuBO73vBpeTv5fWuFfEuZjb7fmSyd4u/mo3jhd4EewTGLNE/By0D2PGupnE7f/M+ij/sfCOa8PAJesJTHLK/JgT1zl87+A L4vFEkVnkF/3nvc5Hwt/L0iSl/NMovr6wMVA2L/fNdcG+qxb21ieSPstrcH19YhTC0/6HEim/coJjbGy36CfZR/6rsM/Rj9+orMG44aFty5voXpabrW8KvDLDKbf1duE9Y5Rq/FHnF3QrFTvBb/nvtC3XQ5+hfPxi6Q/eKlC/qmBxjvY1R7HrbuhR94BjXmymdD3xl132ulcS6L7xB7KX1Qf1MVDvuJf8vdn0HkhceWdXvCr gtytoybkP3/lc2Igdwf1Ccq0ftbl1E5p6HnSK7+9tM/PHUpedIf6Qs5kOw2Hf+L76Wumw6+ZmvKPOcH+1Mb8rtGn8zMOw/ctg50ojHt94RrlvbamQZ/gl3mL8pZ9hF5sFc2L0+F3uPcPTR6kfvct36+/AA/zpve01YBfR2xrMvCA32YWGwweBb4dLA4la1Lf6umed4+Qj2RvfHU9B3g2VfetM4S/4b40cntP5zK89nocojxL HO8uoHrfmmnBO87Bzst+OaSAv2wffXrRSXnNnQ2fJxKOhJ25+Xge3+hCx0E8j/MpZCz14VhoVFnv/AR9t4QkfsG8m38X9h+hOlWZlro99f+khIzXw3xYRjd6Kr8AX0o/X45H/Mx6Ot3j7HPiy4C1oXQOp/W/xhbEwdnyOUwZ8KMpNkj3xf8FnZoHZlP9Q7w3TYHi5cklU6VpH2tfzTJzygMPGvZXQN5Mp1XrLTPoZWKFnAzw wZpbPJMHHmFqXwcnOVL8oDEtk/jT8lXBXuiDfXjaMBfkXYI2h9Tn4BGltsYMe1zP23/11AfoR3jM7sxV6NOhaPrGevBLvO58MwOKFw8u2VxxGfqLWDjzMuYTtF3LbBbyOeZuraUE7KLtk5rZ2be0f7Oq8P960fEJyQeemUAnmdeI61wLdhtTH7e4q/vXI8QzS+ucvDLpvMqB0dIZFJ8oHzedTHXNrZbHT95BPLtth/MgeEPE rp+wHf5Hc+PZcfvonNEry7MOFMePao56Q/srRjO2zaN+jXdGK6xxnVqTv1MHcP5CYC69Avkos4HDmXWwmGm4r+xgRjjL+MmJRt7Flj7xvhB2q5Mw09WT4pkNAjNPioNM0jflYH5b5j4MOkvySJg+TZXq4bljfkRRHsCd6/Me/lXn1ssaAXDgavxbc1EZ7evnuO2l/lurJzGK0HvFTSbZkPrfKtUOqtC+4ZGW4SrUD7HCaFU0 7J3fvOI/oybqu8vU1sW6eSFPt4mAxy1K651WQ65T72t/d8Wnw2N9Vg7ta1nM7NSMIzxpi9cjHh1cMzRiGn43sLrGlsC6uB7Lnc9Tnpe5e3AhnftaOVT9A35elNB6ohjyFNRs2c2CPAuWX69cBX8kbJGX238f/3f8UB5P+6dfnsQ00L51puKGSW7UZzs0ZyHis9Kpq85dAJ9wZpXf/wp/Ihpan1gIPyl7feuOGjwn7O/uI4Uk 186lwjTopUHN7qcJ8bkyN1sL/CTb06Ufg+dyb0UUDdA+/q38EGPYI4cnHSGF+ctuL/L8DLsr4Cik6NE5jXGBj1jkR15mj9tDce/fxCIVqicdsJVbBlx231lqu4fqgHeWmNTjubJTI31MIUeWTVxtLO7zO+vSYwhc110Ka5tI9VCl9fOckvHcivi+PqpbnR8I6IdcRbzLHuWf9JjD881mSFB9YM+XT6eg37CIwsQ/tH9d4azu D1wJwlzGXofceaHFEj/gJzkL2iaMo3w1v8yij/pJPWK3MbTflOfFOwz7FJneOeAM/rSwT4jR6QLe13Ef8cneVxku9KL+SpOu4Fm4viLu5dpc4MevQXX9AJ2nCNvc0Uj7DQsae7XwHM2gJ8orKU6IE8ksuwI/mnlsnyHtE+/7Pt4M85Q95nxiFHDG3hX3YGo6xT1/XLjgkS0TRxgtAw6EkzU1iqhO1nBuUgZd15gTRXGH4KXd 97uYH++jqyrNc0Rnc+qvn8CP39p306gvUFNm8RLoNVLe7Nh42DG7/FxC0G2sb/2q+IkYh/fs05uz+Ax7uKvjIvW1yyRPjKN8obZ33B34X4d3rzfW5FEfW3Z0CnhDquooWxN40Im1O0n7asxpPeOm45iPWfStR4gHzLsOzytFfK+WHyizkHiz8tEB61t03mXCilRaL090IA9+3FQvv9oZOBPdfC4hzqdzNYsbJWPBg2LzGVnQ N/cF25b66bk/DXU8YY+sKf2bVYm/f42acRn25TfcUzQeeJD9LHV4MuxPLfb8sCV4PutN3IoZ4G9RVNY/acRXpZ0HLmwGr1dfaDTVoXp/+oUVGbAPh5a0sYcRl8m+Nlx/FXbDetu59C31RS01NaI6D4+5414IvTbMZFmHwm8zUca+lYgTK7LTjdowPnMkSIbjCH3oPh3IpPg8e3rgT6zDe+pKh6PA8wqdmRKOVCdSiQqfhfiH 07U3b9RT8hONZzZBn6zZm8LvI77LvlhQQfUQvsPt1fMgZ8b106RB+H2FkzrLn9E+80eTyHPhmM/oBbVrrmD86gjbPdCT1PqMbamwM6VPk4wSSS5zFnttA844So+vLc8Fb27eIHwuoHNMnx/5IZ4VtL0bHAn9OMiMGb4B/LX0ywpdP/I3L26eG0d9rocH/o6kfH3pAtMh6psumF/XQXWOuH2HJVOA3wq5EA7iQlvvAl4q1Ued Du+7CD7U0VQ6M5rqOn8XBiRBnhxXTa9q+I/BlPxEL8hfdqegnOoo/HO+/Q8o7g636KFzGOwZv77ZvsX4o9KG+WDcwYx9Rzvb9JihAxULy6lOI/8uKOMu8Przx8I25FnZDaNVZBCHVCQ3KDhR37jy1p2OlA8w5ZlqwVhnH2NmDrtzkKi//RX4zs5YMuF3Ge3LFv0MvoTrAyv7lEiv3H2Nc29gPE3z8hqKj0ad0ZgWSH3BmjYm 1H+hvOqKDezW9eT9cDfyv1vNTslgHQ0q8j+pvqtwwnfKMsi103xHfyP1WZw7GDD8LJ2XsAvvSqbzEmkZ5C+YjiSNa9SPkX5oxlbolbU633k08lVua+v7TuqT5c76/hr+ssBW9s415HmsguI0MeIgzryd4vvkv55cOSuPdQm0nl7ZSfYW7RCpBtxUpCyJ3UD54ZzbEy7RPrrtvOe+4C2O1FvBQ/Bl9NmBGB3q7+Ur7vOhPk87 I4OJiAu3tPyY/ojq2ENDquHAU+emUYYmVF+qrLifSedzHFqjhiDviryJVQ1Y56SRR9x1KG61s5Pl0vo1rmiE4nfXWariDvBJUU3HwbvQn8Lkdk415RccZ6VQ5BW8eP7Hd8D14KQBFSPIQfQqQo5PffffzNasO4fvG5+4CfGd99+DHdnRsAv7/kQb6MUhOWdzFsXNadU2E/NpXzohZzWd7+G/Uo/txDoy/2vZCX8qGB3hIUgr ZuIrNRSXkd8tU3VbTft4XkpvRlH9ZF3z571Yr+0EwZ9owveN0I1bfOGPJpiNbIa8WCcGvA7l0TkcyxjqxxjxcMzhRMLbg+vV6ljX0kn5nwrAk0UWarOCKU4NH69Rieexlz29/Ax5GzuG3VIAPauZ3dlH5+m4l3ZZUJ2LMbO4qovnjJge8HZiB/T5d658AvQeVsv+IEl96dLu8m9CMI98K2E97Wsn+PfmB2D+87+4hSMv0nl9 OozO8WjuFAxW0X6aqn6EFfQhW+O6nkP9AiKz26upXqXbJLoAO3Rv1SzPrsP8gipz4rE+5t2va+2IA7kX1l65BrtM04q+zwGfmVvVNjfA/vmfHt4Ipv2x8zqdtH/FTI9bpsin/gnLx9nwG4LvceqHEQ8YSBbetIed8RTvBs8lP/uC7+IN3uO5dk7pgr7Za0/lbgD+q+W2OJD+2UtDC+jcOPNq8pwOyN9Ua6aBAvFoYD1/CsUr 6suEXVSHG3N4vw/hevHf79G34c/0pna8p3jgwLdfSuCXwXa13/+o7jpvvQn1wVlLN+xeRef5DecEu2N8UU+4+xLIgyM8NFWa8uf9n67up7w3uGRRINbH9Aq+jKRzR/vXxq9FniYW3dq4hM7D7DSUlKD44OyplXZkX9X98hOhr7DKc2p3qU8tuS97H+TCuflDrov6tliJ276m4Hc5l/OO4Fv+s9g7l2n/c47TBjPYwVD02Iv1 iOO3fftcokn7e4o/b4+lvtV6zf5xwIuDPne+B+JLcc99iXkUN9XOsFsCvahN4if20Dma2L5t68AbBqG+G94Sj2n5eG6E/+FeFJwJoHrGtCvZ8y/gOWuausthnyzDZoUd0LumTduoXupjG/0icABx01eZz55r2pEvVz+XuwN+VXihO2wr2fsp+XcyGeQfb6wKpjqv6p9KHvDO1yhfsQK87udWLz2P8KY4c5ce5WWnKo++ozrf w9Iawpvm2Ppcdfqc8+ZvFp3fcP9lxaf9k7maA6epz3XCxNMM5im+tkZVgvLEGIthC6ifUffv2am0P3gnQEeD4tFx09VCsU72iv2Hz0E++o1pV5X+X/9Y8Psi9Cuuz98iAdxEV175Yw68u/62KjtM+Wm48a0e+MuCxSYzzeGv2V2BgX+pv7wqbKEE/Kya+qEddG4ob8m/P1p0vnjZTYNCOg/379uNCj/oq/fW4cRCei/D40R9 qpNvci77Sf1JOst/0XsShHsYt+GYX/ayD6cugXc0xW4Fi8ATaX2rlxPP8pXEWYvhb3LNtKvWUD5yQjf44XXYbZXXyXnUb5F7y8yN6jL936Yep/PErP9OxGCcXWuV23uBuyC/cAcf3Mf6PiraCfmSIDBcLhty5P2t3fwY+lELf1gyuwc49aiesQj873zj+ZwbxB8Pz517aAdellBOM4W8/bbkLk+HvLeYmT00pn6FPO1jbZgv e4KTFPU/dJtHBPpjPIFJsnkI4p3cWe6pBdRvodUo9QT6yZaImHmH9m8Na25YUz5yfDCFzuGwuw68VAF/DB6vfLUW6zbYuC5yD+Su9kXU6foR/nrCoqmjqG+87qFBCZ27WTXl/dEs8MYU6WEu0JOf/q8NG6APzfbR9cthb+Jv3n/CIS8p6Z1uZymeqw1r0oOfy9ZlWr4K//8ehLyD8E8s12+PFsEuuFIRrE7K6+w+fPWH/1U4 v9K8D3KRj5ZMuN5D++zc3Gzih3GV8z5D3qyVz33WAM/cu4ZPzXFfmOPuGqoLqIzO0+SDH6RcrUL+YF2i8Jx8A6qfmLlmLbmJ+fnOrjGhcxG6J9gRlGdeu9H5/3oc38f7bTHk61Dz8wb1P870HLGSzlecXifphnjS+o5Pdxf4gpNpFxEBPYga2R7e1fS+CeWDnVRX8XSXWoP8RSTjmP6E6t+zay5NBn/57f9auJf86/i54+l8 QulrPZGQ6oR5Uu5S0H+d5rEbY8EPg9zT2x9SH6FqaWFALZ2TfeNvS3VPoyL1XsQlnBM3r3yAf4w+kLfjHcbtW3BxjwHw7fxgXHRXP+S5efTghw7I5fTPTgXEscmSIfv8PlJefM7akfJ8+aHkURTnzPnAWoJ5GURptcwjvt85hU/9rIzC+nLFfXRu0OKhlCP1O3L2ZsO/6Ogoxux7TTxgu7MKfkC42eZLFOKnyMdnDvQR71zw kguBH4zOPZ9rCHmW+j0pfI880jZBMtWunt7/cGYk5X2szyYK6sWQ0933vyyoH2KjCaeE6u9jjxdKlVCc8+yC9n3qR93M66Z6BffH+F+4z/6ofZEb9cmavVJ/RXH/6CsO1tT3VzDZzPYw5HjSyWEq+f+FRtuWXYWdPtBO2llVzGio7j3QSOcOPbsMriJeCNPtMHqA/HjXlAcvX9P5lcudH6Ow/i2n9BcvoX2+rWc7TlM8krk3 kueJ8RasLp/0mPoTDv3dRH2Xh2NORiDesJcIKt0B/Rd8mL32JvWHxZiOiIKfjJ7Gyg+F3gzGd9sM0e/O24++QDyS3bv3+w/YkajJafubZKqzFv8LjgHOfkbGydF+9mdnE4cAOr86vT0RelP4zNYOpn0/+zfveYgT9scoXhpH8WCfanEX9cUskusOpfjj0IV8DToXNmrN4CviibFJ7eMeUT9o2X/7cX1ksVvzy/e0v/yldTn8 Kfuay6hgXM+Zq7bdg+retcu2WFK/gbTjcrdQPaZu0aZFTu/gX8tf5pt9gR/MaLGToD66tT2fP2eCD9q776XSvlvXw+kSsHv2y8vhbmHggTGJiYveUF2jt+p7KPQod+zcPtq/P6ZYlO2A/72OaEbcJd5YU9SOeXBkQ66eAX4djFwavlH9p3z4Ck3417RLY+JmUz1Se4ZqDvDO+/i9NC0VevsseXoT8GD6vfTACcqTi5bW3YQd O5zadP0x/JX41H8XV0Nerm0pWT+on/Pn6NE/yd9sedv9EzzIdR9b+wFyFfjvGbDDPAq+7oidi3W5fvzVu57iKYfgjdQnyJx4VqSVAL673vBXkvhU9fy+aOC0YcMS/SrwiMBIPi6I+gI+T9I1wPMdbIXlpZi/w9bMz6Op7rfVfWLfNXxf/+dVMuKs8Vejih7B/kRHnqhqE0/Md5auAC/qb05afZTq1dfuaXrQ+1P+Lba+TXWK 5btbsgsQn1U0THuC+NU5win/J/VRvD62/Qz0Uie6ONoevLbF75umBuQgu3nBlSH4M1eLDXFbKY4LNJ+libjM1szft572AyQN/T2hr3iFm7pc6vuZdm7lm+uQ//zPXadhd9lxbqsH6PzjslMlxvid77NhjhzxxB09H+rHZKw29StCzoNfJkwuB/+lxSfupbiyQej/y4H6y1trVU7S+1he7wx5RvWwDS8+WON+zY4niR9gf2lr CwIOUZ/zwvsW017TezaCs35jvPETb0Q5UF/hlw3vHeEPhFK7dvHp/RTze+p0EZdxZLZ9uwTcFjSlGNaC53m7FaY1wY+JI28//FdFcVezogXygCL93M3vyX68Rn4ogR8QXtZbtP8LeP340Nl3tJ8wdDv0Pr1vZ7aU2AW841qnWp+H8YUBJx9qgW9c/ZdeTSfchNU5pgInFt+meiaS3zbwUU2l+ln62nXKiFOzBfMeHKE6I9vY QIP2NcylfZ3tYR9tMt0LqI/pz3qFPk/gJShb6Ridn5W4enzoKvhvQUWVE/AilrmlOgLzZ0ssujwf/jPsdtSRg9A3uyOzXJ/6cOe17U+huEP3VL4R/LG4ZG/gZepj6mwNLoA/COtuHDYSds22jtrcg3yQb3RvziGqB9jmOVfQ/s5F1X/dsLfctvhLJtTv7LNbeJT2m2L2W9o0Ul1H4/RO2LnQ5smAPNZhWlsWe4HitFwVQWU+ cDTjr8D9Mb3fZMqvPOj75HrrxXV0Hs31sT8f9tVXNiHQDvrasqwmQpb6jA+6Rn3CeHXNm1WmAxfsF7G5xbBPB/fm84HEl38OKCfRuYXRDfaBxEtfdD28YfcNd86HL4B+BedtrsrFYZ49DQYKlPc+aLNtuUt5CuvHniisT4P7dzytP+SN40HEoRUGNbdDEY/oH9Y7+JVwdC5jykqsR3/F/JZ20nOZeLEH7VflJVhGIG8SLtzX mkH29HKYT5gT1qf8uvkq9T2Kjre2Ul4SWN78CXYatmB01maKjyan1NK5Pt27G0vku4qZp9Ovtcnhu/DvZ74x1iH6Ok3rH+J0v1fjtOj9CCz5RQ7dwEFfg3TW11baJ0x5V0n1glbrg08pP3o4S7ya+sW+y2xuigf/jnlVfR/+0mHV4OSx4Em2+9AAnZMfHK03oxj6augsqPlD76eoHDtPi97HpqvOzYJess/nnBoNHhmhv63+ eSedTwiOvQY7DzPdoEH2y7d8o0Tn1C3M43RyP0Gu2Yct31Mde2/AUx3oJ2zwtasN1dUWxuTepP70r/atSpBn9tbPAR50/qxKWXc//BFrxaSCy5ivrPemyFjgrVC33tH+I51re7pVGXjh63Q1DiIOSk47/2Ij9bG3Sg/+oTqcm/BxK9n37/uXzMBDSp5HWjJpH9/URJFD8t29V6UZz9V0ar/5nfo2i86oSYMHxYulZZKfFTO5 PlKPE2mfZa36Uefzxcy0BZzZWgPw1/5DnEng5fhndp30fh/uRrXpVJcSLV8X3wx7E4UXTvKnuv6BymmF1PeRleJ2GXp19b755gJ4mqNyce5izHfp4svu64Bjccyz2g1ltI83YaU/1dct8j6cOYRxv6twpwJH/BvJLEuqR7D0hA6wR+Zw56Mbd6i+IqoZCf2nfTzxNQh61uzsujEJ1wuGXRAMIm61XXLm+C3gunPxo4gVb6jP oTx/CJ/ijpFXJSGHnaPC92z+jnHWCJKK4BfDLo969hL5ot/k5+8PUB9kVNm2rVVUr4oLCwIfsedkbrGGfxLv/BOBGJFhd9Zq/kd1R+15C5Moj20b1aHnDjxmfwp0pXOSh3vbTlD++d1NUhu8vCH8pyCA6lJ9fo3ZufQeidObKum9B+VKlR4YRzNz1Wf1h/QeJJ3ADvC2ztqTBfJkn0JGQkjvrbsi/2AzxYfyT1Pt6X1lrsc1 fWHXnVOje2Po/RCWokWvgUdbmVVee8lvmJ8YdQ/xllqQ1MbQJuDp5DkFek8NJ3VWeSf0JU4z1GFDDtb7M/O/Ud/snHrnnovwOzXtRu2Iv9S0fH1v0rmssJKH4xH38aPP7moguUxpmkb9WYKrAVd6Y/F90SpxOb1fQjlP0E7vqemc6WpLfn2YV2QN+KS7cSd3Nu3zOMtnRFDcnjxCXorO835wKuSDV8KOJP0qAU5to950UV+/ QL3Rx6CW3ld1y24hvb+ntfSHPcVH8Q5n3t3CfPzjB79AXt7ODTW18L/WnDDvZqrvmZQOS6C+xa9XPsy4hXWaaH/yozqjxm2ZYPjP7OmbhxZT/Uu9xiSsms6ZS2hVYV69tabrfIAf9qZGY0mqo39NOrgO8V/09+8q9N4o1jhtj23wX89mR3XTezjUNv16vhzPVdPvDxwH/naIbvg0En6NbXxURoLOU9mt0IkgO369K3gB5V3b z4gpH2VN/yVcQvUHu+RwqpMz2/aqDKP3TK3oqdSCXVgYnRxKx/W2x7rWb4G98E/8LOylcwV7R1n/AL9x7prMpPd55Eoq502m88ezA1QqaB+WYW/wNsNz5D4G0v4q30Vh12hcz9ZqrpyOdXNl/pvwGnzpJ/f3aBL8puZTvU9xwJ1Ip/b0Z/AP56anPu2vcoQt6X+Bcx2h3tatkKf+6u6JM8AzvERFQQHlT6fT+/7yoa+NuWfp nI7fhihHCerb9luZ8A1xjeb+Hp/5FGf+HgjICqFz2Kkjbfxw/9yOvasRL/INlNcfh/8SbX/gvwN2zi9w6KX+Mzi6gHGxtD8iOmpO55qr/IacaD/lv18sN+BJSt/zri78W90hrZOroG/enTwTI8TdpdoztkxAvM9N23TkIPRx8tn2wJZePSago3Usvxd2EbtknzL4nrVp055S2K33yO5Meh9Bg3uUZzrJ6cfFNyrgJ4GkRoEb 5s9+EBnojvENdl1c5offhcd3n9pB71lM9jnxfE8x474x/jOvmfYtLCx3YryCKU1fL1J/2a/s/OMu0MesbxHzkHfYL856afcNzy3JClJCfMx7dfiq30N6D569R08JvZ9KseQ/ynevrcyUiqP3iI06PhU4EsjMnxVO9VrtdhNXf/D4OaPds57Qe7ty5hI+BN1T/pMG3y90W+M09SPWtZ9TNoneO9MVuHddG9Xfsx1/OMBvbRoX zgLPaBaLTsRBvwYal2W6aR+6TcHZhebH1Sl4Cz8mHndC1pTOo3jHadO5n5Na8lsi4acKZrloGNRTfn/mng3FbzE3G24n0vsAzc8UZNB+8dN0W8hHFHS3f91D2v/MKK2g/ZWcakktxPeRd3vveQEXkQ2+k1LBiwbLEmalUZ6/ue9gB+2Tzk/pu4B1hk25mhAFPZu2zW+Rwf+ZjSdcgoE38cq2r6vpvS1BHBt6z57ms2+3G6nf uOftFaoLMN4Kh9osYSelYZIP6PzBmmDmJu1jtjfKD4CPRNx8D1U6D6NU2x8JfxH7/ELG7E7qx1u0Npz8mfGZNweB/7FtKlpdxDePGlrfUBx7LkJ7JZ0bjVztk5yK+9Wk3AwLMe47laoHwJvISfXAXODeYJekYifsW3Of26p7sCehePVFNeCg5tnlwb/kTy7ba99DnMOqZyUm4H6H0on/MnAd5428Hc2X9SWlYS14upMzoyXy K8aza6mJTaP8eooyvV/IwEK9ewU+TX/UNNC5i+ZvD2SUaP+vMjXpPfAnKzVW8g5waRqvq95N53qfJQbRe8EYyWGPdOi86M/5y7cRv5501EygfXkr67X/UqCnT6t3X6c+hjMjeg5j3eoZuZauYuozfqBzHflj2MFnkq30vqfqqqJDkLdG7Lvrs+sonucc0ACObFkxprrAdbYwd9Qw2FWYRO3fabA/vsMunhpwYPrML+A3nSf/ 7XbUAnzJnpijmIHfO18vXeQC/s6eOP3DInofgI5k97WL4JGGJc5D1M+2ZdKiIuqvnTthaoI/1uMv13SkmfoKTqwdTe/Tu6Pu8BTrUdg6Z7dlJZ3bsFr/F/YnlEqQt4d8wxymBt2CXrl58+xHIB7RfHi9+HsBeEwtY2kz6TtWenoA/Puu4sufmsHf0epPC606ad/xuPF8Ot/MzHijRnUQUU/WBap7nttkOfUK7GbWm7vUp8M9 2ahD78UQjTTe2AW75cpnvK7CvEc856ZbUh9X6/p2NazPVu2r5T88jz+7ZMIBxDNid7l3dP6Z0/3K9DXiu0G32OSd1L+xurtzPsXFwZut2oHnAufwdCXqL7P8sicafF0R5V03hd6joFG7nfxV9xgz32d0LvOP969jFCfW7NtYWAw7mnzGuwF68Bs7xfUY8KizXNvRmfoG2heKOolHVwn/bECcJJB9l3qB6sY+eWfHgr+lqgXK xOsi817nSpJ/lt5kDsVXbk9GuQNXSpvc2vaAx+Mdtt6m/eUK1uRhQdQ/dCuRofO/gt/aXZbw91L/6+Daw6HMwvg8bVS6PFhZStKazajUVKpx2TXSMmUwmC4ubZNEJGTk1uYWtbHFiiXVNmQtyiWUZF3WSopkrRVjXB4p19nY9axRK/v+9q/545vvO+e8533PeW+/X6D7An3UoXQNIjTJr5CxmE15E/Rc16HyI9Inxqduux3I zy15+Ni+HvH796tl3unoE2BneJK8WK6HHlXR/xmp7V5OvqSXJayAfNjfAanLSfCIbPCUTYE/k3++UXgPeap5difykwtWTlcCv3cvPnaYxlkmCpDakp4PlAmMqtFP8yxKtg1+43yNpAd8FlrqaQ2ViBfaPu+tBg+uqYkj6osv5dbetK7Alsbgbpwfr0u/DC/AeFZHzEh+DEa97tX1yPM/ZmaiD/yLTaf23qX9CLLwsEde0thT U5CM86dTzAIfmnnU01M0vmSPyZ1q1Mk0em4LKF7iZXk6OtF6J/X882LovtZrN/RIpHlL3pj7+JFfpdhoUvgYeN1EJedlyEtZ8LYdxr280C9FqYd+V1oPxYK3c9XbQTPwCTllV9wivetKnRYmkv9W2qxzuvoNrVOl65Yb3at6I9b7RmieGkXO/Frwqa7n2t7APifnMTbbIY4rKEuie3my5qZvL/lfOr1yW/Ql17EVvn4S0q/W H7xraZ4Dr9y7C4ALDN2voaDnJVluOguRb8o7dtOC5s/tyP6s+AZwJGPu6G+u48nFBx6hz3hJvj/J61utUXuvt+AlPJH0BHwIkVHRgaR/hw5eXM6heUmyrs68wLnrbzVgmgYc1eaxVvC07tWsmkMfp2lfjAnZHefPZvY18stGdIT3L5PemQ5uKUT+3f0db1ITedj3dudRL9JiWXsZ9pNc4/hRB8kfi7gw9lyCOqt4KycE/c4N zUEGpE8lU4n8K7g/Q1S0N5A/xnJUOu5D8d3AlXVnv6N940+7tYejL1g/MSG8jPYvk8GrkIHnLXzCG/14i9Q6diO/qFMkL6bfLqUHVvZ0/ot+mjr3B+lbmMJZPA48p9exuNv3wQ8ut7kGHJDuq/cPSF9G2uI+Xkv72iJVntBV0PeHrZVtwMewzUYYhziqSpgzhnrJdebSWcSdkUIXCfoqCt5Et4LfYtT87Pwl4MSGp8vofAk2 8m5oBy5je3Sm+gv0LWSGniQ7+ll5n34P6UdJZbHJFooT4soOv2snuanmNfXdA+6kq/DrodvQ+7+4O2JJziFncpndqCvNZKuSn5FRGLRzQSvsU17BDqb7zMe2SoXiDoHt30eQx9CTcNUF5LcYp13/oIl6w9GcurBe+u6PG5xMaD/yjGPVAug8VG3JVZtFXNcv5G+MqWVo9cfP8iDPYp2GjXRPDpxx+9UD/BFfOY5yaR5tGptW AA9YriFjnSa7ED0sclEi+29o6k7LoPFOeAbNi4CLszzXuLzv/313NaTx9AJXZDnB/pT3zALXpZqiwkx/Ch5QxWggzStwyu2OA9mRnmDeWZvkNsK9oO5Fz/8pMMpJQZ7c3/pf5mmSa+plh525pFfPO3MW0XdFyw/198EPOz8pXgr5u3J4DsA1LbWzSgWuRLF+LiKb9Px1hMMueo+1aWAuGfU2lYZzEpLLhGORoS94EKztr3hk kH6atebPkZ6I0qJfJuMcKhmRpALPsfugj5jsu04Usvgp9CBsxXUx8i4Zv8VYhgK3tMZAA3WZxbKCXckkN12bGTdaB69EaRn8KUZI2tBx1E0TL51qQF/yaFnII5L7Qp9RYz75zV0rozQy0S/SmVQeWYv+nca14AmVOPFGImHXzd5+2sB1mwWkj+eT3GZq4iton8ttLs0vAV74mza7X+jcHfLcsTmF7K5t7dSFGxhf+5mLPvrt thevWUf+x91VakIj1H/Hj9ZEk1/dtpojeELraWo1sB8mP15vRspNID3SGgwVSFBP3yvYWgn8W3naonngWCOHMxLpHmTri5taSL5hVnFS8GEK5IOt8VLwdQS/jaFzhG+5fdYcvFHHRAkX6T5jZzq7c8oQ9zvlZoF35kOS5SvkzTT7pvbTOlnchHRGPfoROj4pJn+oo1SFo476MofZUkHv7XEtnKqhe3jxeOjv+8hPC6sv75mA vyLtTo+luO5h6Qffi88t/gNfhV85 ```