Automatic merge from submit-queue

StatefulSet: Deflake e2e RunHostCmd.

The initial retry up to 20s was giving up too soon. I'm seeing this test flake because the Node rebooted and it takes ~2min to recover. Now StatefulSet RunHostCmd calls will use the same 5min timeout as with other Pod state checks.

ref #48031