feat(risk): Phase 12 — Risk Intelligence [FASE-12]

- TechniqueRiskProfile model: per-technique risk scoring (0-100)
- 4-factor weighted scoring: detection_gap(35%) + threat_actors(30%) + osint(20%) + test_failures(15%)
- Risk levels: critical(≥75) / high(≥50) / medium(≥25) / low(≥10) / info
- Detailed scoring_breakdown (JSONB) + actionable recommendations per technique
- Router /api/v1/risk: compute-all, compute-one, list, matrix, summary, recommendations, top
- Alembic migration b038risk (raw SQL, idempotent)
- QA script: 60+ tests across all endpoints

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

backend/app/schemas/risk_schema.py

@@ -0,0 +1,71 @@
+"""Phase 12: Risk Intelligence schemas."""
+
+from datetime import datetime
+from typing import Any, Dict, List, Optional
+from uuid import UUID
+
+from pydantic import BaseModel, ConfigDict
+
+VALID_RISK_LEVELS = ["critical", "high", "medium", "low", "info"]
+
+
+class TechniqueRiskProfileOut(BaseModel):
+    model_config = ConfigDict(from_attributes=True)
+
+    id:                 UUID
+    technique_id:       UUID
+    risk_score:         float
+    likelihood:         float
+    impact:             float
+    risk_level:         str
+    detection_gap:      float
+    threat_actor_count: int
+    osint_signal_count: int
+    test_fail_count:    int
+    test_total_count:   int
+    test_failure_rate:  float
+    confidence_level:   float
+    scoring_breakdown:  Optional[Dict[str, Any]]
+    recommendations:    Optional[List[str]]
+    computed_at:        datetime
+    is_stale:           bool
+
+
+class RiskMatrixEntry(BaseModel):
+    model_config = ConfigDict(from_attributes=True)
+
+    technique_id:  UUID
+    technique_name: Optional[str] = None
+    technique_tid:  Optional[str] = None   # e.g. "T1059"
+    risk_score:    float
+    likelihood:    float
+    impact:        float
+    risk_level:    str
+    detection_gap: float
+    computed_at:   datetime
+
+
+class RiskSummary(BaseModel):
+    total_techniques:   int
+    scored_techniques:  int
+    stale_count:        int
+    by_level:           Dict[str, int]       # {"critical": 3, "high": 12, ...}
+    avg_risk_score:     float
+    top_risks:          List[RiskMatrixEntry]
+
+
+class RecommendationItem(BaseModel):
+    technique_id:   UUID
+    technique_name: Optional[str] = None
+    technique_tid:  Optional[str] = None
+    risk_level:     str
+    risk_score:     float
+    recommendations: List[str]
+    priority:       int   # 1 = highest
+
+
+class ComputeResult(BaseModel):
+    computed: int
+    skipped:  int
+    errors:   int
+    duration_seconds: float