Phase 2: Live paper trading engine + extended backtesting analytics

Track A — Live paper trading system:
- Extract PositionManager from backtester into shared src/position_manager.py
- Refactor backtester/engine.py to delegate to PositionManager
- New src/live/ package: data_feed (OANDA polling), executor (paper/live orders),
  engine (LiveEngine orchestrator with 5 strategy slots), run.py entry point
- Add phase2 config to system.yaml (S7_Tight, S9, S9_Filtered, S4F, S3)

Track B — Extended backtesting analytics:
- Regime analysis: per-year (2021-2023) breakdown shows 4/5 strategies trending UP
- Correlation analysis: S7+S3 GBP_JPY overlap=16.9% (moderate), S9 pairs=12% (low)
- Kelly sizing: S9_Filtered half-Kelly=7.3%, S4F=2.4%, S3=1.6% with Monte Carlo DD

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

config/system.yaml

@@ -102,3 +102,30 @@ supabase:
   url: https://<your>.supabase.co
   key_env_name: SUPABASE_KEY
   table: fx_candles
+phase2:
+  strategies:
+    - name: S7_Tight
+      pair: GBP_JPY
+      timeframe: H1
+      enabled: true
+    - name: S9
+      pair: GBP_USD
+      timeframe: H1
+      enabled: true
+    - name: S9_Filtered
+      pair: GBP_AUD
+      timeframe: H1
+      enabled: true
+    - name: S4F
+      pair: EUR_AUD
+      timeframe: M15
+      htf_timeframe: H1
+      enabled: true
+    - name: S3
+      pair: GBP_JPY
+      timeframe: H1
+      enabled: true
+  poll_interval_seconds: 60
+  paper_mode: true
+  starting_equity: 100000
+  max_daily_drawdown_pct: 5.0