4.15 CI Payloads Failing on GCP with Credentials Operator problems

https://prow.ci.openshift.org/view/gs/origin-ci-test/logs/aggregated-gcp-ovn-upgrade-4.15-micro-release-openshift-release-analysis-aggregator/1730445943465054208

https://prow.ci.openshift.org/view/gs/origin-ci-test/logs/aggregated-gcp-ovn-upgrade-4.15-micro-release-openshift-release-analysis-aggregator/1730385372728266752

All jobs that run seem to hit the same quota problem we saw recently:

failed to grant creds: error syncing creds in mint-mode: error creating custom role: rpc error: code = ResourceExhausted desc = Maximum number of roles reached. Maximum is: 300\nerror details: retry in 24h0m1s

This time it seems to be surfacing on a new credentials request from storage: openshift-gcp-pd-csi-driver-operator which was just moved from predefined roles to fine grained permissions in https://github.com/openshift/cluster-storage-operator/pull/410, likely why we're now tripping over this limit.

We're going to revert and buy time for CCO team to investigate.