fix(evoprompt): avoid overflow when counting prompt combinations

[liuzhengyang699]

EvoPrompt counts the prompt combinations with np.prod, so the search-space size can overflow a fixed-width integer even when only a few candidates are requested. With 16 nodes and 16 prompts per node the count wraps to zero and no candidates are produced. With 19 nodes and 10 prompts per node it wraps negative, causing the sampler to enter the full Cartesian-product branch.

Use math.prod to keep the count exact while preserving the existing sampling limits. The regression tests cover both overflow cases with default and explicit sample sizes, as well as small and empty populations. The large-space tests reject full enumeration immediately.

Validation: pytest tests/src/optimizers/test_evoprompt_combinations.py tests/src/optimizers/test_evoprompt_state_isolation.py tests/src/optimizers/test_optimizer_package_imports.py -q — 10 passed. Ruff passed for the new test file; git diff --check passed.

[chatgpt-codex-connector]

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review	Status	Commit	Review trigger
📝 Code Review	✅ Completed 2026-10-10T15:28:18.288792Z	814c9cf	PR opened

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" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.