feat: 全量功能模块开发与集成测试修复

- 新增后端模块：Alert、APIAsset、Compliance、Lineage、Masking、Risk、SchemaChange、Unstructured、Watermark
- 新增前端模块页面与API接口
- 新增Alembic迁移脚本(002-014)覆盖全量业务表
- 新增测试数据生成脚本与集成测试脚本
- 修复metadata模型JSON类型导入缺失导致启动失败的问题
- 修复前端Alert/APIAsset页面request模块路径错误
- 更新docker-compose与开发计划文档

backend/app/tasks/ml_tasks.py

@@ -0,0 +1,26 @@
+from app.tasks.worker import celery_app
+
+
+@celery_app.task(bind=True)
+def train_ml_model_task(self, model_name: str = None, algorithm: str = "logistic_regression"):
+    from app.core.database import SessionLocal
+    from app.services.ml_service import train_model
+
+    db = SessionLocal()
+    try:
+        self.update_state(state="PROGRESS", meta={"message": "Fetching training data"})
+        mv = train_model(db, model_name=model_name, algorithm=algorithm)
+        if mv:
+            return {
+                "status": "completed",
+                "model_id": mv.id,
+                "name": mv.name,
+                "accuracy": mv.accuracy,
+                "train_samples": mv.train_samples,
+            }
+        else:
+            return {"status": "failed", "reason": "Not enough training data (need >= 20 samples)"}
+    except Exception as e:
+        return {"status": "failed", "reason": str(e)}
+    finally:
+        db.close()