fix: log coordinator reload failures in update_gpu_services

- Replace bare `except Exception: pass` with `except Exception as exc` and a
  logger.warning call that surfaces node_id and the exception for diagnostics.
- Move `import os as _os` from mid-file (between test functions) to the
  top-level import block to satisfy PEP 8 and linter expectations.

app/nodes.py

@@ -287,7 +287,7 @@ def update_gpu_services(node_id: str, gpu_id: int, body: UpdateServicesRequest)
                 f"{coordinator_url}/api/nodes/{node_id}/reload-profile", timeout=5.0
             )
             reloaded = rr.status_code < 300
-        except Exception:
-            pass
+        except Exception as exc:
+            logger.warning("Coordinator reload failed for node %s: %s", node_id, exc)
 
     return {"ok": True, "reloaded": reloaded, "warnings": []}

tests/test_nodes.py

@@ -7,6 +7,7 @@ import pytest
 import yaml
 from fastapi.testclient import TestClient
 from unittest.mock import MagicMock, patch
+import os as _os
 
 
 @pytest.fixture(autouse=True)
@@ -202,7 +203,6 @@ def test_get_profile_500_on_malformed_yaml(client, tmp_path):
 
 # ── POST /api/nodes-mgmt/nodes/{node_id}/gpu/{gpu_id}/services ─────────────────
 
-import os as _os
 
 _BASE_PROFILE = {
     "services": {