QTZPOC-20

QTZPOC-20: Refactor model import workflow, introduce atomic model ID generation, and clarify MinIO expiry.

Fuses bundle download, decryption, MLflow experiment creation, and artifact upload into a single activity to prevent cross-worker issues. Implements an atomic counter for MongoDB model IDs, replacing the max(id) + 1 scan to avoid collisions during concurrent imports. Adds comprehensive documentation for MinIO object expiry mechanisms and refines lifecycle rule application. Improves error handling with clearer responsibility classification and centralizes MLflow repository injection.

laborious/activities/activities.py

@@ -20,6 +20,7 @@ with workflow.unsafe.imports_passed_through():
         build_import_config,
         build_import_status_config,
     )
+    from laborious.utils.repository.model_repository import MLFlowRepository
 
 
 class Activities(Storage, MLFlow, Gates, OPC, ModelMetrics, API, ModelImport):
@@ -101,6 +102,15 @@ class Activities(Storage, MLFlow, Gates, OPC, ModelMetrics, API, ModelImport):
             secure=minio_config['secure'],
         )
 
+        mlflow_repository = MLFlowRepository(
+            f'{mlflow_config["host"]}:{mlflow_config["port"]}',
+            mlflow_config['username'],
+            mlflow_config['password'],
+            logger,
+            notification_handler,
+            metrics_controller,
+        )
+
         # Initialize parent classes
         Storage.__init__(
             self,
@@ -120,10 +130,7 @@ class Activities(Storage, MLFlow, Gates, OPC, ModelMetrics, API, ModelImport):
 
         MLFlow.__init__(
             self,
-            mlflow_host=mlflow_config['host'],
-            mlflow_port=mlflow_config['port'],
-            mlflow_username=mlflow_config['username'],
-            mlflow_password=mlflow_config['password'],
+            mlflow_repository=mlflow_repository,
             minio_repository=minio_repository,
             logger=logger,
             notification_handler=notification_handler,
@@ -180,12 +187,7 @@ class Activities(Storage, MLFlow, Gates, OPC, ModelMetrics, API, ModelImport):
             status_config=import_status_config or build_import_status_config(),
             import_config=import_config or build_import_config(),
             minio_repository=minio_repository,
-            # The MLflow repository is shared rather than rebuilt: setting the tracking URI is a
-            # process-wide side effect, so two instances would be two chances to disagree. Read
-            # through `getattr` because `MLFlow.__init__` is what creates it — if that ever stops
-            # running before this line, the import activities refuse with "MLflow repository not
-            # initialized" at the point of use rather than silently provisioning nothing.
-            mlflow_repository=getattr(self, 'model_monitoring_repository', None),
+            mlflow_repository=mlflow_repository,
             mongo_repository=mongo_repository,
             logger=logger,
             notification_handler=notification_handler,