Wire C30 peer returns into production + add C29 MetaLabel log

Merge C19: adaptive triple-barrier labeling into old_wang_dev

Add C19 adaptive triple-barrier labeling (PASSED 5-stock and 12-stock)

Integrate Old Wang trading strategy overlay

Backup current stock predictor strategies

Apply optimized factor weight overlay

Temporarily skip login and add validation config report

feat: integrate local architecture with HF Space

fix(C7): revert calendar features from FEATURE_COLUMNS — up_precision regressed -5.8pp

feat(C7): add calendar/seasonality features + 36m lookback; PASSED

feat(C5): SHAP pruning — remove 11 zero-contribution features, +1pp metrics

feat(C6): add HMM regime detection — FAILED (improvement <1.5pp aggregate)

feat(C4): add meta-labeling secondary classifier — PASSED (up_prec 51.7→55.8%)

feat(C3): add triple barrier labels — PASSED (dir_acc 31.4→41.2%, up_prec 43.0→51.7%)

feat: feature audit + 3-way backtest, update FEATURE_COLUMNS to winner subset

feat: Phase 4 — securities lending, block trades, large holder features

feat: add CatBoost to prediction ensemble with balanced class weights

feat: add SOX and TNX cross-asset features

feat: add TAIFEX put/call ratio and large trader features

feat: add Amihud illiquidity, momentum features, and 200MA regime gate

feat: Phase 1 prediction improvements — alignment, BIAS, interactions, confidence gate

feat: add chip_score, MA120/240, margin flow, and TA distillation

feat: add institutional signals and HF-friendly deploy

feat: ML ensemble upgrade — LightGBM, cross-asset features, walk-forward embargo

feat(ml): Phase 3 — sentiment proxy + auto-threshold + cross-asset features

feat(ml): Optuna walk-forward tuning + /tune endpoint

feat(ml): XGBoost ensemble + dynamic thresholds for accuracy improvement

fix(metal): Phase 3 — import fix + predictor threshold kwargs + smoke test

fix: handle low-data stocks gracefully (009816, new ETFs)

feat: comprehensive optimization — 30 improvements across ML, perf, frontend, security, architecture

perf: eliminate redundant _build_features in train/predict pipeline

fix: remove hardcoded creds, add rate limiting, unify thresholds, cache stock info

refactor: improve ML features, LINE Flex Messages, watchlist panel, health endpoint

fix+refactor: LINE webhook key bug, fetcher DRY, feature importance, async predictor service

feat: add buy_target price to prediction panel

deploy: initial release