feat: Phase 4+5 — price discovery status + replay historique

Phase 4 — Price Discovery Status (la pièce maîtresse) :
- price_discovery.py (nouveau) : capture_price_snapshots() sauve les prix des tickers
  liés à chaque news scorée (energy→BZ=F/NG=F, metals→GC=F/HG=F, indices→^GSPC/IWM)
- compute_absorptions() mesure combien du mouvement attendu s'est déjà produit
  (status: not_yet_priced <30% / partially_priced 30-80% / fully_priced >80%)
- build_price_discovery_block() → bloc prompt avec opportunités classées
- database.py : table news_price_snapshots + save/get/purge fonctions
- auto_cycle.py : capture après ai_score_news_batch, compute avant suggestions,
  block injecté dans suggestion + scoring prompts + context snapshot
- ai_analyzer.py : param price_discovery_block dans suggest + score

Phase 5 — Replay historique :
- cycle.py : POST /api/cycle/contexts/{run_id}/replay — recharge le snapshot historique
  et relance suggest_patterns_from_market_context avec le contexte original
- useApi.ts : hook useReplayCycle
- SystemLogs.tsx : bouton "Rejouer ce cycle" dans onglet Contexte IA avec champ
  notes, résultats inline (liste des patterns générés), section price_discovery
  ouverte par défaut en rouge

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

backend/routers/cycle.py

@@ -98,3 +98,96 @@ def get_context_snapshot(run_id: str):
     if not snap:
         raise HTTPException(404, "Snapshot non trouvé pour ce cycle")
     return snap
+
+
+class ReplayRequest(BaseModel):
+    override_notes: Optional[str] = None  # optional annotation added to the replay
+
+
+@router.post("/contexts/{run_id}/replay")
+def replay_cycle(run_id: str, req: ReplayRequest):
+    """
+    Phase 5 — Re-run AI suggestion with the historical context from a saved snapshot.
+    Returns new pattern suggestions based on the original context data.
+    """
+    snap = get_cycle_context_snapshot(run_id)
+    if not snap:
+        raise HTTPException(404, "Snapshot non trouvé pour ce cycle")
+
+    ctx = snap.get("context", {})
+    if not ctx:
+        raise HTTPException(400, "Snapshot vide — impossible de rejouer")
+
+    try:
+        from services.ai_analyzer import suggest_patterns_from_market_context, apply_news_decay, partition_news_by_age
+        import json
+
+        # Reconstruct minimal inputs from the snapshot
+        # news: merge inter_cycle + recent_24h + older from partitioned snapshot
+        news_part = ctx.get("news_partitioned", {})
+        news_flat = (
+            news_part.get("inter_cycle", [])
+            + news_part.get("recent_24h", [])
+            + news_part.get("older", [])
+        )
+
+        # quotes: reconstruct from quotes_summary
+        quotes_by_class = ctx.get("quotes_summary", {})
+
+        # calendar
+        calendar = ctx.get("calendar", [])
+
+        # macro regime (simplified for replay)
+        macro_regime = None
+        if ctx.get("macro_regime"):
+            macro_regime = {"scenarios": {"dominant": ctx["macro_regime"].get("dominant"), "scores": ctx["macro_regime"].get("scores"), "asset_bias": {}, "reasons": []}, "gauges": {}}
+
+        # geo score
+        geo_score = ctx.get("geo_score", {"score": 50, "level": "medium", "top_risks": []})
+
+        # preserved blocks
+        tech_block = ctx.get("tech_indicators_block", "")
+        iv_context = ctx.get("iv_context_preview", "")
+        cycle_meta = ctx.get("cycle_meta", {})
+
+        # Rebuild FRED block from saved releases
+        fred_block = ""
+        if ctx.get("fred_releases"):
+            from services.fred_fetcher import build_fred_context_block
+            fred_block = build_fred_context_block(ctx["fred_releases"])
+
+        # Rebuild price discovery block from saved absorptions
+        pd_block = ""
+        if ctx.get("price_discovery"):
+            from services.price_discovery import build_price_discovery_block
+            pd_block = build_price_discovery_block(ctx["price_discovery"])
+
+        # Add replay note to cycle_meta
+        if req.override_notes:
+            cycle_meta = {**cycle_meta, "replay_notes": req.override_notes}
+        cycle_meta["is_replay"] = True
+        cycle_meta["replayed_at"] = __import__("datetime").datetime.utcnow().isoformat()
+
+        suggestions = suggest_patterns_from_market_context(
+            news=news_flat,
+            quotes_by_class=quotes_by_class,
+            calendar=calendar,
+            macro_regime=macro_regime,
+            geo_score=geo_score,
+            iv_context=iv_context,
+            cycle_meta=cycle_meta,
+            tech_indicators_block=tech_block,
+            fred_block=fred_block,
+            price_discovery_block=pd_block,
+        )
+
+        return {
+            "run_id": run_id,
+            "original_ts": snap["ts"],
+            "replayed_at": cycle_meta["replayed_at"],
+            "override_notes": req.override_notes,
+            "suggestions_count": len(suggestions),
+            "suggestions": suggestions,
+        }
+    except Exception as e:
+        raise HTTPException(500, f"Replay failed: {str(e)}")