ci: remove scheduled OpenTelemetry runs

[zhongkechen]

Summary

• remove the daily schedule from the OpenTelemetry conformance workflow
• keep pull request, push, and manual dispatch triggers
• update the workflow contract test to prevent scheduled runs from being reintroduced

This stops automatic runs such as https://github.com/aws/aws-durable-execution-sdk-python/actions/runs/34323141348.

Testing

• 15 GitHub script tests passed, including test_opentelemetry_conformance_workflow.py

[github-actions]

Claude AI review

This PR only touches CI configuration: it removes the schedule: trigger (daily cron) from .github/workflows/opentelemetry-conformance-tests.yml and updates .github/scripts/tests/test_opentelemetry_conformance_workflow.py to assert that no schedule: block is present, replacing the previous assertion that a specific cron string existed.

Verification performed:

• Diff is limited to the two files listed in the PR metadata (additions: 1, deletions: 3), matching the stated intent.
• The remaining on: triggers (pull_request, push, workflow_dispatch) are unchanged, so per-PR/push/manual runs are preserved as claimed.
• The updated test assertion (assert "\n schedule:" not in workflow) is a stronger, more future-proof check than the old exact-cron-string match — it will fail if any schedule: trigger (with any cron expression) is reintroduced, not just the specific "0 7 * * *" one. This matches the stated goal of "prevent scheduled runs from being reintroduced."
• No core SDK, testing, OTel, or examples package code is touched, so there is no impact on checkpoint/replay determinism, durable operation semantics, serialization, or public API surfaces.
• The change follows the repo's existing pattern of string-based workflow-contract tests (already used for other assertions in the same file), so it is consistent with house style; it doesn't introduce a new fragility pattern.
• No unsafe YAML loading or other security-sensitive parsing is introduced.

No correctness, regression, determinism, concurrency, or API-compatibility issues were found. Residual risk is limited to the pre-existing style of asserting on raw YAML text rather than parsed YAML, which could in theory miss a differently-indented reintroduction of a schedule trigger, but this matches the existing test suite's established pattern and is not a new risk introduced by this PR.

Reviewed commit 22bee7eac7a6801c2f29e17e07002e9cee5cf01d. Workflow run

[github-actions]

Codex AI review

No actionable findings. Residual risk is low; the workflow trigger change was reviewed statically and not exercised in GitHub Actions.

Reviewed commit 22bee7eac7a6801c2f29e17e07002e9cee5cf01d. Workflow run