Use most powerful Linux runner for publishing

[ammario]

Summary

• use the most powerful Linux runner when publishing

Testing

• cargo test (fails: tests::test_concurrent_jail_isolation, tests::test_jail_allows_matching_requests, tests::test_jail_denies_non_matching_requests, tests::test_jail_dns_resolution, tests::test_jail_exit_code_propagation, tests::test_jail_https_connect_denied, tests::test_jail_method_specific_rules, tests::test_jail_network_diagnostics, tests::test_jail_privilege_dropping, tests::test_jail_request_log, tests::test_namespace_cleanup, tests::test_native_jail_blocks_https, tests::test_outbound_tcp_non_http_blocked)

---

https://chatgpt.com/codex/tasks/task_e_68c1d54cb11083298f9c120c40ae18d8

[chatgpt-codex-connector]

Codex Review: Here are some suggestions.

Reply with @codex fix comments to fix any unresolved comments.

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, or 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 fix this CI failure" or "@codex address that feedback".

[chatgpt-codex-connector]

[P1] Avoid unavailable GitHub runner label

Switching the publish job to runs-on: ubuntu-latest-xlarge assumes the repository has access to GitHub’s larger hosted runners. Unless the organization is explicitly configured for large runners, this label does not resolve and the workflow fails to start with “No hosted runner found” errors, blocking every release. Consider retaining ubuntu-latest or using a self-hosted label that is guaranteed to exist.

Useful? React with 👍 / 👎.