feat(api): Phase 15.2 Redis Trace Context 傳遞

實現 Redis Streams 跨服務追蹤零斷鏈:
- telemetry.py: 新增 get_trace_context() + restore_trace_context()
- webhooks.py: Producer 注入 _trace_id, _span_id 到 Redis
- signal_worker.py: Consumer 還原 Trace Context 建立子 Span

架構: API → Redis Streams → Worker 完整追蹤鏈
格式: W3C Trace Context (traceparent)

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

apps/api/src/workers/signal_worker.py

@@ -2,18 +2,25 @@
 Signal Worker - Redis Streams Consumer
 =======================================
 Phase 6.1: Event Bus Implementation
+Phase 15.2: Redis Trace Context Propagation (2026-03-26)
 
 功能:
 - XREADGROUP 消費 stream:awoooi_signals
 - Signal → Incident 聚合邏輯 (Phase 6.3 實作)
 - 失敗重試 + ACK 機制
 - Graceful Shutdown
+- **Phase 15.2**: Trace Context 還原 (零斷鏈觀測)
 
 Redis Streams 概念:
 - Stream: stream:awoooi_signals (訊息佇列)
 - Consumer Group: awoooi_workers (消費者群組)
 - Consumer: worker_{hostname} (單一消費者)
 
+Trace Context 傳遞 (Phase 15.2):
+- Producer (webhooks.py) 寫入 _trace_id, _span_id 到 Redis
+- Consumer (此檔案) 還原 Context，建立子 Span
+- 實現 API → Redis → Worker 完整追蹤鏈
+
 統帥鐵律:
 - 使用 XREADGROUP 確保訊息只被處理一次
 - 處理完成後必須 XACK
@@ -27,6 +34,7 @@ from typing import Any
 import structlog
 
 from src.core.redis_client import get_redis, get_worker_redis
+from src.core.telemetry import restore_trace_context
 from src.services.incident_engine import get_incident_engine
 
 logger = structlog.get_logger(__name__)
@@ -195,57 +203,96 @@ class SignalWorker:
         3. Incident 建立/更新 (聚合到同一 Incident)
         4. GraphRAG 爆炸半徑分析
         5. 雙層持久化 (Redis + PostgreSQL)
+
+        Phase 15.2: Trace Context 還原
+        - 從 Redis 訊息提取 _trace_id, _span_id
+        - 建立子 Span 繼承原始 Trace，實現零斷鏈觀測
         """
         redis_client = get_redis()
 
-        try:
-            logger.info(
-                "signal_received",
-                message_id=message_id,
-                source=data.get("source", "unknown"),
-                alert_name=data.get("alert_name", "unknown"),
-                severity=data.get("severity", "unknown"),
-                namespace=data.get("namespace", "default"),
-                target=data.get("target", "unknown"),
-            )
+        # =================================================================
+        # Phase 15.2: 提取 Trace Context (從 Producer 注入的欄位)
+        # =================================================================
+        trace_context = None
+        trace_id = data.pop("_trace_id", None)  # pop 避免污染 signal data
+        span_id = data.pop("_span_id", None)
 
-            # Phase 6.3: 使用 IncidentEngine 處理訊號
-            # - 自動聚合相關告警到同一 Incident
-            # - GraphRAG 分析爆炸半徑
-            # - 雙層持久化
-            engine = get_incident_engine()
-            incident = await engine.process_signal(data)
+        if trace_id:
+            trace_context = {
+                "trace_id": trace_id,
+                "span_id": span_id or "",
+            }
+
+        # =================================================================
+        # 在還原的 Trace Context 中處理訊號
+        # =================================================================
+        with restore_trace_context(trace_context) as span:
+            try:
+                # 設置 Span 屬性 (用於 SignOz 搜尋)
+                span.set_attribute("messaging.system", "redis_streams")
+                span.set_attribute("messaging.destination", STREAM_KEY)
+                span.set_attribute("messaging.message_id", message_id)
+                span.set_attribute("signal.source", data.get("source", "unknown"))
+                span.set_attribute("signal.alert_name", data.get("alert_name", "unknown"))
+                span.set_attribute("signal.severity", data.get("severity", "unknown"))
 
-            if incident:
                 logger.info(
-                    "signal_processed_by_engine",
+                    "signal_received",
                     message_id=message_id,
-                    incident_id=incident.incident_id,
-                    severity=incident.severity.value,
-                    signal_count=len(incident.signals),
-                    affected_services=incident.affected_services,
-                    persisted_to_pg=incident.persisted_to_pg,
-                )
-            else:
-                logger.warning(
-                    "signal_processing_failed",
-                    message_id=message_id,
-                    signal_data=data,
+                    source=data.get("source", "unknown"),
+                    alert_name=data.get("alert_name", "unknown"),
+                    severity=data.get("severity", "unknown"),
+                    namespace=data.get("namespace", "default"),
+                    target=data.get("target", "unknown"),
+                    trace_restored=trace_context is not None,
                 )
 
-            # ACK: 確認訊息已處理
-            await redis_client.xack(STREAM_KEY, CONSUMER_GROUP, message_id)
+                # Phase 6.3: 使用 IncidentEngine 處理訊號
+                # - 自動聚合相關告警到同一 Incident
+                # - GraphRAG 分析爆炸半徑
+                # - 雙層持久化
+                engine = get_incident_engine()
+                incident = await engine.process_signal(data)
 
-            logger.debug("signal_acked", message_id=message_id)
+                if incident:
+                    # 記錄 Incident 到 Span
+                    span.set_attribute("incident.id", incident.incident_id)
+                    span.set_attribute("incident.severity", incident.severity.value)
+                    span.set_attribute("incident.signal_count", len(incident.signals))
 
-        except Exception as e:
-            logger.exception(
-                "signal_process_error",
-                message_id=message_id,
-                error=str(e),
-            )
-            # 不 ACK，訊息會留在 Pending List
-            # Phase 6.3 將實作 Pending List 清理機制
+                    logger.info(
+                        "signal_processed_by_engine",
+                        message_id=message_id,
+                        incident_id=incident.incident_id,
+                        severity=incident.severity.value,
+                        signal_count=len(incident.signals),
+                        affected_services=incident.affected_services,
+                        persisted_to_pg=incident.persisted_to_pg,
+                    )
+                else:
+                    span.set_attribute("signal.processing_failed", True)
+                    logger.warning(
+                        "signal_processing_failed",
+                        message_id=message_id,
+                        signal_data=data,
+                    )
+
+                # ACK: 確認訊息已處理
+                await redis_client.xack(STREAM_KEY, CONSUMER_GROUP, message_id)
+                span.set_attribute("messaging.acked", True)
+
+                logger.debug("signal_acked", message_id=message_id)
+
+            except Exception as e:
+                span.set_attribute("error", True)
+                span.set_attribute("error.message", str(e))
+                logger.exception(
+                    "signal_process_error",
+                    message_id=message_id,
+                    error=str(e),
+                )
+                # 不 ACK，訊息會留在 Pending List
+                # Phase 6.3 將實作 Pending List 清理機制
 
 
 # =============================================================================