Notify Telegram on runtime entry failures

[Pigbibi]

Summary

• add a top-level runtime exception fallback for LongBridge HTTP entrypoints
• send Telegram alerts for failures that escape normal strategy-cycle handling
• return HTTP 500 when the strategy reports a failed run without throwing
• cover the fallback and failure-response paths with request handling tests

Tests

• /usr/bin/python3 -m py_compile main.py tests/test_request_handling.py
• git diff --check
• /usr/bin/python3 -m unittest tests.test_request_handling -q

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 2bbbf5375f

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[chatgpt-codex-connector]

Keep handled strategy failures acknowledged

When run_strategy() catches a strategy_cycle exception it now returns False, and this branch converts that handled failure into an HTTP 500 for the Cloud Scheduler-backed /, /backfill, and /precheck routes. In scheduler jobs with retries enabled, a broker/strategy error that may have occurred after submitting part of an order cycle is now treated as an unacknowledged request and can rerun the same cycle; previously these handled failures were notified via publish_cycle_notification but the route still acknowledged the trigger. Reserve the 500/runtime fallback for exceptions that escape before the strategy-cycle handler runs.

Useful? React with 👍 / 👎.