That is, if we haven't merged a flake.lock PR by the time the job runs again, the job should push to the existing branch, updating the unmerged PR with the latest flake.lock changes.
Currently it fails because the branch already exists; instead of failing, it should push to the existing branch.
That is, if we haven't merged a
flake.lock
PR by the time the job runs again, the job should push to the existing branch, updating the unmerged PR with the latestflake.lock
changes.Currently it fails because the branch already exists; instead of failing, it should push to the existing branch.