Bump task_timeout on debug-mode test to tolerate CI contention

The test_workflow_runs_on_main_thread_in_debug_mode test passes locally
in 0.3s against time-skipping, localhost dev-server, and under -n auto
parallelism, but hangs in CI against the bundled dev server. Most
likely the server's default workflow task timeout (10s) is being raced
by inline-activation latency under heavy CI parallel load, causing
task reassignment and a stuck client awaiting completion.

Set task_timeout=timedelta(minutes=2) on the workflow as a conservative
temporary mitigation while we diagnose. If CI goes green with this
change the theory holds. If still red, the issue is elsewhere and this
should come back out.

Author: elidlocke

tests/worker/test_breakpoint_hang.py

@@ -17,6 +17,7 @@
 import sys
 import threading
 import uuid
+from datetime import timedelta
 
 from temporalio import workflow
 from temporalio.client import Client
@@ -78,6 +79,12 @@ async def test_workflow_runs_on_main_thread_in_debug_mode(client: Client):
             ThreadCaptureWorkflow.run,
             id=f"wf-{uuid.uuid4()}",
             task_queue=task_queue,
+            # Conservative task_timeout to tolerate CI contention. The inline
+            # activation blocks the asyncio loop briefly; under heavy parallel
+            # CI load this can race the server's default workflow task timeout
+            # and cause task reassignment. Temporary mitigation while we
+            # diagnose the CI-specific hang against the bundled dev server.
+            task_timeout=timedelta(minutes=2),
         )
 
     main_name = threading.main_thread().name