docs(reflex): link AlignmentBench from Alignment use case

[0xnavarro]

Public change

Links the new public AlignmentBench developer preview from Reflex and the Reflex Alignment use case. The docs now distinguish target-model behavioral evaluation (AlignmentBench), decision-engine evaluation (ReflexBench), and execution-time semantic supervision (Reflex Alignment).

Validation

• I used only public/synthetic data.
• I did not copy private Brida material.
• Public-surface checks pass.
• No secrets, internal URLs/IDs or customer data are included.
• This PR does not weaken security/review policy.
• Commits are narrowly scoped and atomic.

Review / QA evidence

Ran pnpm check after regenerating focused Reflex skills. Registry validation passes, 44 focused skills are synchronized, 27/27 tests pass, and the full-history public leak scan passes.

Compatibility / risk

Documentation/generated-skill update only. No recipe schema, hosted route, authorization, metering or execution authority changes.

Release intent

• release-impact: none
• release-note: none
• qa-scope: Run registry validation, focused-skill synchronization, the full test suite and the public-surface leak scan.

[chatgpt-codex-connector]

Codex Review Summary

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

Review	Status	Commit	Review trigger
🔒 Security Review	✅ Completed 2026-09-23T20:32:33.687751Z	26c7845	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.