Pass reserved cash policy to strategy runtime

[Pigbibi]

Summary

• pass positive reserved cash floor/ratio into strategy runtime overrides
• update UsEquityStrategies dependency to 0cd90a37 with cash-reserve floor support
• add regression coverage for runtime reserve policy overrides

Tests

• PYTHONPATH=. timeout 180 /tmp/qsl-followup-pytest-venv-20260619/bin/python -m pytest tests/test_strategy_runtime.py::test_reserved_cash_policy_overrides_apply_to_runtime_config
• git diff --check

[chatgpt-codex-connector]

💡 Codex Review

FirstradePlatform/pyproject.toml

Line 18 in b49064b

"us-equity-strategies @ git+https://github.com/QuantStrategyLab/UsEquityStrategies.git@361338f60900182e3be535cd5fd2be2b9a07b422",

Keep the UsEquityStrategies pins in sync

This pin now differs from requirements.txt line 5, while CI and Docker install from requirements.txt and the README quick start installs the package via pip install -e . from pyproject.toml. That means local/package installs exercise UsEquityStrategies@361338..., but tested and deployed images use @0cd90..., so strategy-runtime behavior such as the new cash-reserve support can diverge; pin both manifests to the same commit.

---

FirstradePlatform/runtime_config_support.py

Lines 177 to 178 in b49064b

	dca_mode=_optional_dca_mode_env("DCA_MODE"),
	dca_base_investment_usd=_optional_positive_float_env("DCA_BASE_INVESTMENT_USD"),

Propagate the DCA env vars in Cloud Run sync

For Cloud Run deployments managed by .github/workflows/sync-cloud-run-env.yml, these new settings are never supplied because that workflow only exposes and passes the allowlisted variables around lines 48-112 and 512-572, and it does not include DCA_MODE or DCA_BASE_INVESTMENT_USD. As a result, setting those repository vars will still leave deployed DCA profiles on defaults; add the two vars to the workflow env and add_optional_env list.

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".