short kubectl timeout to avoid race condition grabbing jobs

This should fix the race condition where packages fire up a job which is running as the wait script started. Jobs finish and complete with ready:false, so the prior implementation would timeout and fail.

This was almost always seen now with weave and kyverno-policies. This pipeline shows the fix working: https://repo1.dso.mil/big-bang/pipeline-templates/package-validation/-/pipelines/2045794