Machine config operator does not upgrade non-master nodes when node roles are customized

Issue

• When upgrading the cluster, machine config operator reports a successful upgrade for the master pool, but not the worker pool:

$ oc get mcp
NAME     CONFIG                                             UPDATED   UPDATING   DEGRADED   MACHINECOUNT   READYMACHINECOUNT   UPDATEDMACHINECOUNT   DEGRADEDMACHINECOUNT
master   rendered-master-7351f04453991cf43b2c1471541a013d   True      False      False      3              3                   3                     0
worker   rendered-worker-2ce490d2dc745e30b8249f16371e57e2   True      False      False      0              0                   0                     0

• Custom node roles are in place.