T211702, T211703, and this task all seem slightly at odds with each other. If we clone everything before running quibble, then we lose the opportunity to optimize for the fast-fail on linting. If CI runs npm and composer test, then quibble shouldn't be doing the cloning, etc. We need to refine this plan before implementing, IMHO.

Would it be better to feed a list of desired actions into a single quibble job, then have that work in stages (repo first, then dependencies)? Or is there some advantage to running separate jobs, e.g. easier to read the reporting and artifacts? Are there jobs other than quibble which should also run on the cloned environment, e.g. phan, and do these require the isolated repo, or all dependencies? If job can run in parallel, then it makes sense for the cloning step to happen in a separate job. Unless we're considering building some of the parallelism into quibble?