Fleet-wide CI-speed pass: masterplan-lock.yml had no job-level
timeout-minutes (a hung runner could block indefinitely) and triggers
on both push (branches: ["**"]) and pull_request, which runs the same
check twice per PR commit. Add a 10-minute timeout for this cheap
single-check job and a concurrency group (matching the pattern already
used in admin/serv0, admin/s0cial, admin/ppl0, admin/pers0n) to cancel
the superseded duplicate run without changing what triggers CI.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
