Fix MLflow logged-models API compat in pipeline tutorial

The curated sklearn-1.5 environment ships MLflow 2.19+ which uses the
logged-models API not supported by AzureML tracking server, causing
mlflow.sklearn.log_model() and mlflow.register_model() to crash.

- autolog(log_models=False) to prevent autolog hitting the API
- Replace log_model() with save_model + log_artifacts + MlflowClient
  registration which use supported endpoints

Author: Chakradhar886

tutorials/get-started-notebooks/pipeline.ipynb

@@ -560,7 +560,7 @@
     "mlflow.start_run()\n",
     "\n",
     "# enable autologging\n",
-    "mlflow.sklearn.autolog()\n",
+    "mlflow.sklearn.autolog(log_models=False)\n",
     "\n",
     "os.makedirs(\"./outputs\", exist_ok=True)\n",
     "\n",
@@ -609,11 +609,15 @@
     "\n",
     "    # Registering the model to the workspace\n",
     "    print(\"Registering the model via MLFlow\")\n",
-    "    mlflow.sklearn.log_model(\n",
-    "        sk_model=clf,\n",
-    "        registered_model_name=args.registered_model_name,\n",
-    "        artifact_path=args.registered_model_name,\n",
-    "    )\n",
+    "    mlflow.sklearn.save_model(sk_model=clf, path=\"model\")\n",
+    "    mlflow.log_artifacts(\"model\", artifact_path=args.registered_model_name)\n",
+    "    run_id = mlflow.active_run().info.run_id\n",
+    "    client = mlflow.tracking.MlflowClient()\n",
+    "    try:\n",
+    "        client.create_registered_model(args.registered_model_name)\n",
+    "    except mlflow.exceptions.MlflowException:\n",
+    "        pass\n",
+    "    client.create_model_version(name=args.registered_model_name, source=f\"runs:/{run_id}/{args.registered_model_name}\", run_id=run_id)\n",
     "\n",
     "    # Saving the model to a file\n",
     "    mlflow.sklearn.save_model(\n",
@@ -946,4 +950,4 @@
  },
  "nbformat": 4,
  "nbformat_minor": 1
-}
+}