refactor(api): Phase 17 P1 Tier 3 紅區服務 Protocol 定義

新增 5 個紅區核心服務的 Protocol 介面：
- IDecisionManager: 決策狀態機
- ITrustScoreManager: 信任評分引擎
- IIncidentEngine: 事件處理引擎
- IMultiSigRedisService: 分散式鎖服務
- ITelegramSecurityInterceptor: 安全攔截器

符合 leWOOOgo 積木化規範：
- 支援依賴注入 (DI)
- 便於測試時 Mock
- 型別約束確保實作一致性

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

apps/api/src/services/trust_engine.py

@@ -19,7 +19,7 @@ Phase 3.2: Progressive Autonomy
 import logging
 from dataclasses import dataclass, field
 from datetime import datetime
-from typing import Literal
+from typing import Literal, Protocol, runtime_checkable
 
 # Phase 16 R2 (2026-03-25): RiskLevel 統一改從 models/approval.py 導入
 # 原因: 消除重複定義，統一風險等級來源
@@ -107,6 +107,57 @@ class TrustThresholds:
 DEFAULT_THRESHOLDS = TrustThresholds()
 
 
+# ==================== Protocol Interface (Phase 17 P1) ====================
+
+
+@runtime_checkable
+class ITrustScoreManager(Protocol):
+    """
+    TrustScoreManager 介面定義
+
+    用途:
+    - 依賴注入 (DI) 時的型別約束
+    - 測試時 Mock 的型別檢查
+    - 符合 leWOOOgo 積木化規範
+
+    Tier 3 紅區服務: 修改需首席架構師簽核
+
+    @see feedback_lewooogo_modular_enforcement.md
+    @see docs/RED_ZONES.md
+    """
+
+    def record_approval(
+        self,
+        action_pattern: str,
+        user_role: str,
+        user_id: str | None = None,
+    ) -> TrustRecord:
+        """記錄人類批准"""
+        ...
+
+    def record_rejection(
+        self,
+        action_pattern: str,
+        user_role: str,
+        user_id: str | None = None,
+        reason: str | None = None,
+    ) -> TrustRecord:
+        """記錄人類拒絕"""
+        ...
+
+    def evaluate_adjusted_risk(
+        self,
+        action_pattern: str,
+        original_risk: str | RiskLevel,
+    ) -> RiskAdjustment:
+        """評估調整後的風險等級"""
+        ...
+
+    def get_trust_record(self, action_pattern: str) -> TrustRecord | None:
+        """取得信任記錄"""
+        ...
+
+
 # ==================== Trust Engine ====================