ci: compare native Bun scheduling with the pinned baseline

[mldangelo-oai]

Summary

The runner experiment used a different Bun baseline from required CI and mixed version changes with execution-mode changes. Make the native scheduling comparison explicit before considering replacement of the custom dispatcher.

Changes

• Align the experimental baseline with required Bun 1.3.14 and select Bun 1.4.0 as a separate candidate, with each version pinned once.
• Compare the baseline with the unsharded candidate, then compare that candidate with three native shards on Linux and seven on Windows.
• Convert the existing per-platform CI estimates to Bun's native timing format. Keep the same seed and compare test identities and outcomes, including failures and missing reports.
• Focus this optional workflow on native sharding; it no longer runs the old isolation and process-parallel trials. Machine-wide policy tests remain excluded from the experiment and run in their existing serial CI step.
• Update two test-side terminal assumptions: strip dashboard frame escapes with Bun's terminal helper, and allow either benign C1 payload behavior while retaining the control-character, credential, and unsafe-link assertions. Production sanitization is unchanged.
• Use native file URLs in dashboard assertions and JSON formatting for multiline test parameters. This preserves the assertions and gives both Bun versions stable test identities in comparison reports.

Testing

• Focused comparison: 101 cases passed on both versions; three native weighted shards produced the same 101-case inventory and outcomes.
• Portability follow-up: all 313 release and dashboard cases passed on both versions with identical test identities and outcomes.
• Full pinned SDK suite with seed 12345: 2,072 passed, 39 skipped.
• Local unsharded Bun 1.4 trial after the terminal test updates: 2,072 passed, 39 skipped.
• Typechecking and formatting passed.
• Full pinned randomized SDK suite with seed 3341195835: 2,072 passed, 39 skipped.

Risk and rollout

This is an experimental workflow change. Required CI, its Bun version, and its custom shard dispatcher remain unchanged. Native process parallelism and isolation are not enabled. The comparison gathers compatibility and inventory evidence; it does not promote the candidate or claim a CI speedup.

Public disclosure review

• No customer, partner, prospect, or user identities, data, or identifying details are included.
• No credentials, personal data, private source, scan findings, or nonpublic links or tickets are included.
• I reviewed the branch name, title, description, commits, changes, comments, logs, screenshots, attachments, and links for public disclosure.

[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 2026-08-30T02:39:36.352763Z	90fac1e	Manual request
🔒 Security Review	✅ Completed 2026-08-30T02:39:21.636632Z	90fac1e	New commits

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.

[mldangelo-oai]

@codex review

Please review the current head, 2602515.

[chatgpt-codex-connector]

Codex Review: Didn't find any major issues. Chef's kiss.

Reviewed commit: 26025153fc

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".

[mldangelo-oai]

@codex review

Please review the current head, 90fac1e.

[chatgpt-codex-connector]

Codex Review: Didn't find any major issues. More of your lovely PRs please.

Reviewed commit: 90fac1e680

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".