Fix resource completeness script after helm install

The script waits for the helm install job to complete:

$ kubectl -n magnum-tiller wait job/install-cern-magnum-job --for=condition=complete --timeout=600s || jobstatus=1

There seems to be 2 issues with the script:

1. It waits for 600s while helm install has configurable timeouts, with the default being 15m (https://gitlab.cern.ch/kubernetes/magnum/-/blob/cern/train/magnum/drivers/common/templates/kubernetes/helm/cern-chart.sh#L376)
2. In case the wait above times out, job status is set to 1, but the value is not used afterwards. The script enters a loop that waits for all daemonsets to be ready.

I suspect we got into point 1. in INC3809327, where the helm install timeout after 15m, but the cluster is in CreateComplete state (the cluster is 1.29).