Add multi-factor conviction gate to activation

Make "qualified" mean an edge candidate, not just R:R + confidence.
The gate now also requires (all admin-configurable, defaults on):
- high conviction: recommended_action LONG_HIGH / SHORT_HIGH only
- clean read: risk_level Low (no contradicting signals)
- probable primary target: best target probability >= min (default 60)

- Shared predicate: app/services/qualification.py +
  frontend/src/lib/qualification.ts (mirrored)
- Activation config extended (min_target_probability,
  require_high_conviction, exclude_conflicts) with bool-aware
  get/update + validation
- /trades/performance switched to ?qualified_only=true, applying
  the full gate server-side; confidence breakdown stays unfiltered
- Dashboard "Qualified", Signals "Qualified only" toggle, and
  Track Record all use the one gate; Admin gains the new controls

Sentiment provider runtime config (prior change) included.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

app/schemas/admin.py

@@ -59,6 +59,21 @@ class TickerUniverseUpdate(BaseModel):
 
 
 class ActivationConfigUpdate(BaseModel):
-    """Activation thresholds: what counts as an actionable signal."""
+    """Activation gate: what counts as an actionable signal."""
     min_rr: float | None = Field(default=None, ge=0)
     min_confidence: float | None = Field(default=None, ge=0, le=100)
+    min_target_probability: float | None = Field(default=None, ge=0, le=100)
+    require_high_conviction: bool | None = None
+    exclude_conflicts: bool | None = None
+
+
+class SentimentConfigUpdate(BaseModel):
+    """Runtime sentiment LLM config. api_key is write-only; omit/empty to keep
+    the stored key."""
+    provider: Literal["openai", "gemini"] | None = None
+    model: str | None = Field(default=None, max_length=100)
+    api_key: str | None = Field(default=None, max_length=400)
+
+
+class SentimentTestRequest(BaseModel):
+    ticker: str = Field(default="AAPL", max_length=10)