```

refactor(Dockerfile): 使用requirements.txt替代硬编码依赖

将Dockerfile中的硬编码pip包列表替换为通过requirements.txt文件安装，
提高依赖管理的灵活性和可维护性。

feat(scheduling): 移除内置APScheduler，采用独立调度系统

移除app/core/background/scheduler.py中原来的APScheduler实现，
改为使用新的应用级调度系统app.application.scheduling。

refactor(task_queue): 调整任务队列模块结构和导入路径

将任务队列相关常量从app.core.background.task_queue.constants迁移至
app.tasks.constants，并更新所有相关导入路径和引用。

refactor(events): 将事件服务重构为独立的应用层组件

将事件分发逻辑从核心层迁移到应用层，使用app.application.events.EventDispatchService
替代原有的app.modules.events.services.EventService。

feat(ai_memory): 增强AI记忆自动写入的安全策略

新增ai_memory_blocked_content_terms配置项用于阻止敏感内容，
添加TTL过期机制控制自动写入条目的生命周期。

fix(security): 强化生产环境安全验证机制

增加model_validator确保生产环境中数据库连接、API密钥、CORS设置等
关键安全配置符合要求。

feat(risks): 优化风险事件操作动作的外键约束

为RiskEventAction模型的风险事件ID字段添加外键约束，
防止孤立记录并增强数据完整性。

refactor(audit): 优化审计服务方法命名和事务处理

将AuditService的log方法重命名为record以反映其阶段行为，
并调整事务提交时机以提高性能。

feat(events): 增强领域事件并发处理和响应模型

添加事件锁定机制防止重复处理，更新API响应模型以提供
更准确的数据类型定义。
```

app/application/pipelines/lifecycle.py

@@ -0,0 +1,237 @@
+from datetime import date
+from typing import Any
+
+from sqlalchemy import select
+from sqlalchemy.orm import Session
+
+from app.core.config import get_settings
+from app.core.security import business_mutations_enabled
+from app.core.constants import ActorValue
+from app.core.utils.time import utc_now
+from app.application.delivery import ReportDeliveryService
+from app.modules.feishu.service import FeishuService
+from app.modules.legacy_mysql.intasect import IntasectSyncService
+from app.modules.reports.constants import ReportPushStatus, ReportType
+from app.modules.reports.services import ReportService
+from app.modules.workflows.constants import WorkflowStatus, WorkflowType
+from app.modules.workflows.models import WorkflowInstance
+from app.modules.workflows.service import WorkflowService
+
+
+class LifecyclePipelineService:
+    def __init__(self, db: Session):
+        self.db = db
+        self.workflows = WorkflowService(db)
+
+    def period_key(self, report_type: str, reference_date: date | None = None) -> str:
+        start, end = ReportService(self.db)._management_period(report_type, reference_date)
+        period_value = end.isoformat() if report_type == ReportType.DAILY else start.isoformat()
+        return f"{report_type}:{period_value}"
+
+    def find(self, period_key: str) -> WorkflowInstance | None:
+        return self.db.execute(
+            select(WorkflowInstance).where(
+                WorkflowInstance.workflow_type == WorkflowType.LIFECYCLE_REPORT,
+                WorkflowInstance.aggregate_type == "report_period",
+                WorkflowInstance.aggregate_id == period_key,
+            )
+        ).scalar_one_or_none()
+
+    def prepare(
+        self,
+        report_type: str,
+        actor: str,
+        force: bool = False,
+    ) -> tuple[WorkflowInstance, str, bool]:
+        period_key = self.period_key(report_type)
+        existing = self.find(period_key)
+        if existing is not None and existing.status == WorkflowStatus.COMPLETED and not force:
+            return existing, period_key, True
+        workflow = self.workflows.start_or_update(
+            workflow_type=WorkflowType.LIFECYCLE_REPORT,
+            aggregate_type="report_period",
+            aggregate_id=period_key,
+            status_value=WorkflowStatus.RUNNING,
+            action="queued",
+            actor=actor,
+            payload={"report_type": report_type, "period_key": period_key, "force": force},
+        )
+        return workflow, period_key, False
+
+    def run(
+        self,
+        report_type: str,
+        receive_id: str | None = None,
+        receive_id_type: str = "chat_id",
+        force: bool = False,
+        actor: str = ActorValue.SCHEDULER,
+    ) -> dict[str, Any]:
+        if not business_mutations_enabled():
+            return {
+                "period_key": self.period_key(report_type),
+                "deduplicated": False,
+                "status": "operations_disabled",
+            }
+        workflow, period_key, deduplicated = self.prepare(report_type, actor, force)
+        if deduplicated:
+            return {
+                "workflow_code": workflow.code,
+                "period_key": period_key,
+                "deduplicated": True,
+                "status": workflow.status,
+            }
+        try:
+            workflow = self.workflows.start_or_update(
+                workflow_type=WorkflowType.LIFECYCLE_REPORT,
+                aggregate_type="report_period",
+                aggregate_id=period_key,
+                status_value=WorkflowStatus.RUNNING,
+                action="source_sync",
+                actor=actor,
+                payload={"report_type": report_type},
+            )
+            sync_result = IntasectSyncService(self.db).sync_all(
+                run_code=workflow.code,
+                force_full=False,
+            )
+            self.workflows.start_or_update(
+                workflow_type=WorkflowType.LIFECYCLE_REPORT,
+                aggregate_type="report_period",
+                aggregate_id=period_key,
+                status_value=WorkflowStatus.RUNNING,
+                action="analysis",
+                actor=actor,
+                payload={
+                    "datasets": {name: result["processed"] for name, result in sync_result.items()}
+                },
+            )
+            report_service = ReportService(self.db)
+            report = report_service.management_lifecycle_report(
+                report_type=report_type,
+                actor=actor,
+                include_ai=True,
+            )
+            settings = get_settings()
+            target_receive_id = receive_id or settings.feishu_default_chat_id
+            ai_analysis = report.get("ai_analysis") or {}
+            if not ai_analysis.get("ok"):
+                notified = self._notify_ai_unavailable(
+                    target_receive_id,
+                    receive_id_type,
+                    period_key,
+                    actor,
+                )
+                failed = self.workflows.start_or_update(
+                    workflow_type=WorkflowType.LIFECYCLE_REPORT,
+                    aggregate_type="report_period",
+                    aggregate_id=period_key,
+                    status_value=WorkflowStatus.FAILED,
+                    action="ai_unavailable",
+                    actor=actor,
+                    payload={
+                        "ai_unavailable": True,
+                        "notified": notified,
+                        "error_type": ai_analysis.get("type") or "AIUnavailable",
+                    },
+                )
+                return {
+                    "workflow_code": failed.code,
+                    "period_key": period_key,
+                    "deduplicated": False,
+                    "status": failed.status,
+                    "ai_unavailable": True,
+                    "notified": notified,
+                }
+            idempotency_key = period_key
+            if force:
+                idempotency_key = f"{period_key}:force:{utc_now():%Y%m%d%H%M%S%f}"
+            push_run = report_service.create_push_run(
+                report_type=report_type,
+                title=str(report["title"]),
+                receive_id=target_receive_id,
+                receive_id_type=receive_id_type,
+                actor=actor,
+                status=ReportPushStatus.PENDING,
+                idempotency_key=idempotency_key,
+            )
+            if push_run.status != ReportPushStatus.SUCCESS:
+                ReportDeliveryService(self.db).push_report(
+                    report,
+                    target_receive_id,
+                    receive_id_type,
+                    actor,
+                    push_run_code=push_run.code,
+                )
+            completed = self.workflows.start_or_update(
+                workflow_type=WorkflowType.LIFECYCLE_REPORT,
+                aggregate_type="report_period",
+                aggregate_id=period_key,
+                status_value=WorkflowStatus.COMPLETED,
+                action="pushed",
+                actor=actor,
+                payload={"push_run_code": push_run.code, "idempotency_key": idempotency_key},
+            )
+            return {
+                "workflow_code": completed.code,
+                "period_key": period_key,
+                "push_run_code": push_run.code,
+                "deduplicated": False,
+                "status": completed.status,
+            }
+        except Exception as exc:
+            self.db.rollback()
+            self.workflows.start_or_update(
+                workflow_type=WorkflowType.LIFECYCLE_REPORT,
+                aggregate_type="report_period",
+                aggregate_id=period_key,
+                status_value=WorkflowStatus.FAILED,
+                action="failed",
+                actor=actor,
+                payload={"error": str(exc)[:2000]},
+            )
+            self._notify_failure(receive_id, receive_id_type, period_key, exc, actor)
+            raise
+
+    def _notify_ai_unavailable(
+        self,
+        receive_id: str | None,
+        receive_id_type: str,
+        period_key: str,
+        actor: str,
+    ) -> bool:
+        settings = get_settings()
+        if (
+            not receive_id
+            or not settings.feishu_app_id
+            or not settings.feishu_app_secret
+        ):
+            return False
+        FeishuService(self.db).send_text(
+            f"生命周期报告 {period_key}：AI 当前不可用，本次分析报告未发送。请检查模型服务。",
+            receive_id,
+            receive_id_type,
+            actor,
+        )
+        return True
+
+    def _notify_failure(
+        self,
+        receive_id: str | None,
+        receive_id_type: str,
+        period_key: str,
+        error: Exception,
+        actor: str,
+    ) -> None:
+        settings = get_settings()
+        target = receive_id or settings.feishu_default_chat_id
+        if not target or not settings.feishu_app_id or not settings.feishu_app_secret:
+            return
+        try:
+            FeishuService(self.db).send_text(
+                f"生命周期报告 {period_key} 执行失败：{type(error).__name__}",
+                target,
+                receive_id_type,
+                actor,
+            )
+        except Exception:
+            self.db.rollback()