ryanrhughes merged
shbernal's omacom/omarchy#16045 into quattro. The fix keeps a plain Bash session leader above the updater so terminal loss ends the waiting prompt and releases its lock. A recorded process ID helps diagnose contention, while flock remains the authority for locking.
tobi reported an abandoned updater that had survived nearly seven days, then updated the contributor branch and its tests through maintainer edits. In omacom/omarchy#16045:pullrequestreview-5906052520,
tobi reports that ordinary terminal closure passed on the unchanged base, while abruptly losing the script process stranded real gum; the patch passed both cases. This reproduces the observed process state without establishing why script originally disappeared.
