Add Firstrade strategy precheck runtime reports

[Pigbibi]

Summary

• wrap Firstrade strategy runs with runtime_report.v1 persistence
• make /precheck run a silent dry-run strategy cycle instead of a health-only response
• suppress cycle notifications and plugin alert dispatch during Firstrade precheck
• copy plugin summaries and plugin errors into the runtime report shape used by other platforms

Config

• set EXECUTION_REPORT_GCS_URI=gs://qsl-runtime-logs-shared/execution-reports for FirstradePlatform

Tests

• git diff --check
• python -m py_compile main.py application/rebalance_service.py tests/test_request_handling.py
• PYTHONPATH=/home/ubuntu/Projects/QuantPlatformKit/src timeout 120 python -m pytest tests/test_request_handling.py -q (skipped locally: Flask is not installed in this VPS env)

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 50f9914475

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]

Disable precheck strategy-state persistence

When FIRSTRADE_PERSIST_STRATEGY_RUNS=true, forcing dry_run_only here still leaves persist_strategy_runs enabled for /precheck. run_strategy_cycle then writes a dry-run latest.json to the same strategy-run key used by live idempotency, and application/strategy_run_persistence.py:is_duplicate_live_run ignores dry_run_only states, so a scheduled precheck after a live terminal run can make the next live /run in that period submit again instead of being skipped as a duplicate. Disable strategy-run state persistence for forced prechecks while still persisting the runtime report.

Useful? React with 👍 / 👎.

[chatgpt-codex-connector]

Use unique IDs for runtime reports

This run ID only has one-second resolution, but persist_runtime_report derives the local/GCS report path from run_id. If Cloud Run handles two requests for the same strategy/account in the same second (for example a scheduler retry overlapping a manual /run, with concurrency enabled), the later report overwrites the earlier one, losing one runtime record. Include a unique suffix such as a UUID or request ID in addition to the timestamp.

Useful? React with 👍 / 👎.