Fix training log race conditions and audit event error

Critical fixes for training session logging:

1. Training log race condition fix:
   - Add explicit session commits after creating training logs
   - Handle duplicate key errors gracefully when multiple sessions
     try to create the same log simultaneously
   - Implement retry logic to query for existing logs after
     duplicate key violations
   - Prevents "Training log not found" errors during training

2. Audit event async generator error fix:
   - Replace incorrect next(get_db()) usage with proper
     async context manager (database_manager.get_session())
   - Fixes "'async_generator' object is not an iterator" error
   - Ensures audit logging works correctly

These changes address race conditions in concurrent database
sessions and ensure training logs are properly synchronized
across the training pipeline.

services/training/app/api/training_operations.py

@@ -236,27 +236,27 @@ async def start_training_job(
 
         # Log audit event for training job creation
         try:
-            from app.core.database import get_db
-            db = next(get_db())
-            await audit_logger.log_event(
-                db_session=db,
-                tenant_id=tenant_id,
-                user_id=current_user["user_id"],
-                action=AuditAction.CREATE.value,
-                resource_type="training_job",
-                resource_id=job_id,
-                severity=AuditSeverity.MEDIUM.value,
-                description=f"Started training job (tier: {tier})",
-                metadata={
-                    "job_id": job_id,
-                    "tier": tier,
-                    "estimated_dataset_size": estimated_dataset_size,
-                    "quota_usage": quota_result.get('current', 0) if quota_result else 0,
-                    "quota_limit": quota_limit if quota_limit else "unlimited"
-                },
-                endpoint="/jobs",
-                method="POST"
-            )
+            from app.core.database import database_manager
+            async with database_manager.get_session() as db:
+                await audit_logger.log_event(
+                    db_session=db,
+                    tenant_id=tenant_id,
+                    user_id=current_user["user_id"],
+                    action=AuditAction.CREATE.value,
+                    resource_type="training_job",
+                    resource_id=job_id,
+                    severity=AuditSeverity.MEDIUM.value,
+                    description=f"Started training job (tier: {tier})",
+                    metadata={
+                        "job_id": job_id,
+                        "tier": tier,
+                        "estimated_dataset_size": estimated_dataset_size,
+                        "quota_usage": quota_result.get('current', 0) if quota_result else 0,
+                        "quota_limit": quota_limit if quota_limit else "unlimited"
+                    },
+                    endpoint="/jobs",
+                    method="POST"
+                )
         except Exception as audit_error:
             logger.warning("Failed to log audit event", error=str(audit_error))