fix(codex): preserve sequential reasoning summaries

[riverpilot]

Summary

• Preserve Codex reasoning-summary deltas when sequential-cutoff delivery is requested.
• Prefer atomic .done summaries when available and flush buffered deltas at output-item completion.
• Add regression coverage for delta-only sequential delivery.

Screenshots

No visual change.

Testing

• bun test packages/ai/test/openai-codex-responses-lite.test.ts packages/ai/test/openai-codex-stream.test.ts packages/ai/test/openai-codex.test.ts
• bun run --cwd=packages/ai check:types
• Live OMP Codex smoke with --print-thoughts completed in the source worktree.

Risk

• Limited to Codex reasoning-summary stream assembly.
• Does not expose private raw chain-of-thought; it preserves summaries emitted by the API.

Notes

• OMP continues requesting reasoning.summary = "detailed" when thinking is enabled.

AI Review Report

The change is narrowly scoped to sequential summary event delivery and includes a delta-only regression test.

Security Audit

No authentication, authorization, secret handling, or external request policy changes.

[chatgpt-codex-connector]

Codex usage limits have been reached for code reviews. Please check with the admins of this repo to increase the limits by adding credits.
Repo admins can enable using credits for code reviews in their settings.