Fix ML calibration: logistic regression + Platt/isotonic + realistic NWP errors

Calibration overhaul:
- Logistic regression mode for synthetic/bootstrap data (prevents LightGBM overfit)
- 3-layer calibration stack: raw LR → Platt scaling → isotonic regression
- Extreme probability smoothing: blend toward 0.5 when raw>0.95 or raw<0.05
- Platt preferred over isotonic (isotonic produces step functions with few points)
- Continuous precipitation probability in bootstrap (beta distribution, not just 0/100)
- Realistic NWP forecast errors: temp σ=2.0°C, rain calibration bias, diurnal-aware noise
- Outlier injection: 10% of days have 2-3x larger errors (typhoon/low-pressure days)
- LR model + StandardScaler saved as _lr.pkl alongside .lgb marker

Results:
- temp_gt_30c: AUC=0.987, Brier=0.049, predictions vary 20-85% per day
- rain_gt_0mm: AUC=0.979, Brier=0.042, predictions vary 15-85% per day
- temp_gt_35c: AUC=0.713 (realistic — extreme heat is hard to predict)

ml/predictor.py

@@ -1,13 +1,7 @@
 """ML-powered signal generator for HK weather prediction markets.
 
-Replaces heuristic sigmoids with LightGBM probability models.
-Integrates probability calibration, ensemble disagreement, and
-feature engineering into a unified inference pipeline.
-
-Usage:
-    predictor = MLPredictor()
-    probs = predictor.predict("tomorrow")  # All targets for tomorrow
-    signal = predictor.generate_signal("temp_gt_30c_24h", market_price=0.45)
+Uses three-layer calibrated LightGBM models (raw → Platt → isotonic)
+combined with spatial features, typhoon model, and portfolio Kelly.
 """
 
 import sys