fix: read actualPhase from type-specific workload endpoints

The /workloads/distributed/{id} and /workloads/trainings/{id} endpoints
return actualPhase, not phase (which was the field on the generic
/workloads/{id} endpoint). This caused a KeyError crash immediately
after the 403 fix landed.

Now reads actualPhase first, falls back to phase for compatibility,
and returns None (PENDING) if neither field is present.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Signed-off-by: oliver könig <okoenig@nvidia.com>

Author: ko3n1g

nemo_run/core/execution/dgxcloud.py

@@ -377,7 +377,11 @@ def status(self, job_id: str) -> Optional[DGXCloudState]:
             return None
 
         r_json = response.json()
-        return DGXCloudState(r_json["phase"])
+        phase = r_json.get("actualPhase") or r_json.get("phase")
+        if not phase:
+            logger.warning(f"No phase field in status response for job {job_id}: {r_json}")
+            return None
+        return DGXCloudState(phase)
 
     def fetch_logs(
         self,

test/core/execution/test_dgxcloud.py

@@ -877,7 +877,7 @@ def test_nproc_per_node_default(self):
     def test_status(self, mock_get):
         mock_response = MagicMock()
         mock_response.status_code = 200
-        mock_response.json.return_value = {"phase": "Running"}
+        mock_response.json.return_value = {"actualPhase": "Running"}
         mock_get.return_value = mock_response
 
         with patch.object(DGXCloudExecutor, "get_auth_token", return_value="test_token"):
@@ -903,7 +903,7 @@ def test_status(self, mock_get):
     def test_status_distributed(self, mock_get):
         mock_response = MagicMock()
         mock_response.status_code = 200
-        mock_response.json.return_value = {"phase": "Running"}
+        mock_response.json.return_value = {"actualPhase": "Running"}
         mock_get.return_value = mock_response
 
         with patch.object(DGXCloudExecutor, "get_auth_token", return_value="test_token"):