feat(p3.1-t2): Tier-2 三服務感知強化 — Sentry 簽章 + DiagnosisAggregator + Solver actions test

Wave 8 P3.1-T2 三項感知強化（多 engineer 補完）：

Sentry Webhook 簽章驗證:
- sentry_webhook.py: 接入 SentryWebhookService.verify_sentry_signature()
- 拒絕無效 sentry-hook-signature → 401 → 防偽造攻擊

DiagnosisAggregator Pod 深診斷整合:
- pre_decision_investigator.py: 新增 _collect_diagnosis_aggregator()
- ENABLE_DIAGNOSIS_AGGREGATOR feature flag 守衛（default=False）
- evidence_snapshot.py: extra_diagnosis 欄位 + build_summary 顯示
- timeout=3.0s + try/except 隔離（fail-soft）
- Conservative 策略：待重疊分析確認 vs PreDecisionInvestigator 不重複

config.py:
- 新增 ENABLE_DIAGNOSIS_AGGREGATOR Field（default=False，K8s ConfigMap 動態啟用）

Solver B1 補測（commit 7c726ebc 對應）:
- test_solver_recommended_actions.py — 20 tests + 3 skipped
- 驗證結構化 recommended_actions（北極星 §1.1 修復多樣性 ≥ 40%）
- LLM 失敗 graceful degraded（candidates=[], degraded=True）

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Co-Authored-By: Multiple Engineers (Wave 8 P3.1-T2) <noreply@anthropic.com>

apps/api/src/services/evidence_snapshot.py

@@ -92,6 +92,8 @@ class EvidenceSnapshot:
     # Phase 4 ADR-084: 動態異常感官（DynamicBaseline + LogAnomaly + TrendPredictor）
     # 2026-04-15 ogt + Claude Sonnet 4.6（亞太）: Phase 4 8D 升級
     anomaly_context: dict[str, Any] | None = None    # Phase 4 動態異常上下文
+    # 2026-04-27 P3.1-T2 by Claude — DiagnosisAggregator Pod 深診斷補充（in-memory only，不持久化）
+    extra_diagnosis: str | None = None
 
     # 感官品質
     mcp_health: dict[str, bool] = field(default_factory=dict)
@@ -162,6 +164,9 @@ class EvidenceSnapshot:
             parts.append(f"[依賴拓撲] {self.dependency_topology}")
         if self.anomaly_context:
             parts.append(f"[動態異常偵測]\n{self.anomaly_context}")
+        # 2026-04-27 P3.1-T2 by Claude — DiagnosisAggregator Pod 深診斷（ENABLE_DIAGNOSIS_AGGREGATOR=true 時填入）
+        if self.extra_diagnosis:
+            parts.append(f"[Pod深診斷]\n{self.extra_diagnosis}")
 
         # 感官品質報告
         failed_tools = [t for t, ok in self.mcp_health.items() if not ok]

apps/api/src/services/pre_decision_investigator.py

@@ -149,6 +149,21 @@ class PreDecisionInvestigator:
         except Exception:
             logger.exception("phase4_anomaly_collect_error", incident_id=incident_id)
 
+        # 4.6 P3.1-T2 by Claude 2026-04-27 — DiagnosisAggregator Pod 深診斷（守門：ENABLE_DIAGNOSIS_AGGREGATOR）
+        # Conservative 策略：預設關閉，避免與 MCP sensor 重複收集 K8s+SignOz 資料。
+        # 待重疊分析完成確認互補性後，由統帥設定 ENABLE_DIAGNOSIS_AGGREGATOR=true 啟用。
+        try:
+            from src.core.config import settings as _settings
+            if _settings.ENABLE_DIAGNOSIS_AGGREGATOR:
+                await asyncio.wait_for(
+                    self._collect_diagnosis_aggregator(snapshot, incident),
+                    timeout=3.0,
+                )
+        except asyncio.TimeoutError:
+            logger.warning("diagnosis_aggregator_collect_timeout", incident_id=incident_id)
+        except Exception:
+            logger.warning("diagnosis_aggregator_collect_failed", incident_id=incident_id)
+
         # 5. 組裝 summary
         snapshot.evidence_summary = snapshot.build_summary()
 
@@ -171,6 +186,48 @@ class PreDecisionInvestigator:
         )
         return snapshot
 
+    async def _collect_diagnosis_aggregator(
+        self,
+        snapshot: EvidenceSnapshot,
+        incident: "Incident",
+    ) -> None:
+        """
+        P3.1-T2 by Claude 2026-04-27 — DiagnosisAggregator Pod 深診斷整合
+
+        僅在 ENABLE_DIAGNOSIS_AGGREGATOR=true 時呼叫（外層已守門）。
+        從 incident labels 取 pod_name + namespace，呼叫 DiagnosisAggregator
+        收集 K8s events + SignOz metrics，結果存入 snapshot.extra_diagnosis。
+
+        Conservative 策略說明：
+        DiagnosisAggregator 與 MCP sensors（D1_K8S_STATE / D3_METRICS）存在資料重疊，
+        本方法透過 feature flag 隔離，不影響主路徑。資料僅作補充，不覆蓋 MCP 結果。
+        """
+        from src.services.diagnosis_aggregator import get_diagnosis_aggregator
+
+        labels = _get_labels(incident)
+        pod_name = labels.get("pod", labels.get("name", ""))
+        namespace = labels.get("namespace", "awoooi-prod")
+
+        if not pod_name:
+            logger.debug("diagnosis_aggregator_skip_no_pod", incident_id=snapshot.incident_id)
+            return
+
+        aggregator = get_diagnosis_aggregator()
+        ctx = await aggregator.collect_pod_diagnosis(
+            pod_name=pod_name,
+            namespace=namespace,
+        )
+        prompt_ctx = ctx.get_llm_prompt_context()
+        if prompt_ctx:
+            snapshot.extra_diagnosis = prompt_ctx[:4000]  # 限 4K chars，不壓縮主 evidence_summary
+            logger.debug(
+                "diagnosis_aggregator_collected",
+                incident_id=snapshot.incident_id,
+                pod=pod_name,
+                signals=len(ctx.signals),
+                highest_severity=ctx.highest_severity.value,
+            )
+
     async def _collect_phase4_anomalies(self, snapshot: EvidenceSnapshot) -> None:
         """
         Phase 4 8D 感官增強：從 ProactiveInspector 快取 + LogAnomalyDetector