chore(e2e): add logs to determine source of failure

[Hweinstock]

Problem

The strands py container test is failing in CI only (passes locally) and we are missing sufficient logs to determine the source:

 FAIL  e2eTest/project/templates.test.ts > add, dev, deploy, invoke for runtime templates > local invocation > 'py_strands_container' runs locally
Error: Test timed out in 180000ms.
If this is a long-running test, pass a timeout value as the last argument or configure it globally with "testTimeout".
 ❯ e2eTest/project/templates.test.ts:236:35
    234|       });
    235|
    236|       test.each(RUNTIME_TEMPLATES)(
       |                                   ^
    237|         "$name runs locally",

    238|         { concurrent: true, timeout: TIMEOUT_MS.PROJECT_INVOKE },

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯[1/1]⎯

 Test Files  1 failed (1)
      Tests  1 failed | 37 passed (38)
   Start at  20:43:12
   Duration  1084.32s (transform 58ms, setup 27.28s, import 94ms, tests 1056.80s, environment 0ms)

Solution

Add some error details to be able to root cause the issue.

[agentcore-devx-automation]

Claude Security Review: no high-confidence findings. (run)

[agentcore-devx-automation]

AgentCore Harness Review

Verdict: Looks good

Small, focused diagnostic PR that adds error logs before each throw inside the retry block in e2eTest/project/templates.test.ts, plus a lightweight TestLogger helper. Purpose (root-causing the py_strands_container CI-only failure) is clear and the change is scoped to e2e tests only. No production code, no telemetry surface, and no new mocks.

One non-blocking observation the author may want to consider as a follow-up: the logger.error(pendingOutput) calls sit inside the retry(...) callback, which retries every 250ms up to TIMEOUT_MS.PROJECT_INVOKE. In particular the !port branch is the expected state while the dev server warms up, so on a slow CI run this will emit the full pending output on every attempt — potentially a lot of duplicated log noise. Logging only the final failure (e.g. by having retry surface the last error to the caller, or by moving the diagnostic dump into a catch around the retry call) would give the same signal with less noise. Fine to land as-is for the immediate debugging need.

[codecov-commenter]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 97.25%. Comparing base (bafd09e) to head (84d7c2b).

Additional details and impacted files

@@            Coverage Diff            @@
##           refactor    #2394   +/-   ##
=========================================
  Coverage     97.25%   97.25%           
=========================================
  Files           613      613           
  Lines         41028    41028           
=========================================
  Hits          39902    39902           
  Misses         1126     1126           

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[agentcore-devx-automation]

Claude Security Review: no high-confidence findings. (run)

[agentcore-devx-automation]

Claude Security Review: no high-confidence findings. (run)