Nodepool metric does not correctly reflect nodepool state

hypershift_nodepools_available_replicas does not properly reflect the nodepool.

$ oc get nodepools -n ocm-production-12345678
NAME              CLUSTER   DESIRED NODES   CURRENT NODES   AUTOSCALING   AUTOREPAIR   VERSION   UPDATINGVERSION   UPDATINGCONFIG   MESSAGE
re-test-workers   re-test   2               0               False         True         4.12.35                                      Minimum availability requires 2 replicas, current 0 available

Meanwhile, there are 3 hypershift_nodepools_available_replicas time series for the nodepools:
- re-test-worker2 reporting 1
- re-test-worker3 reporting 1
- re-test-workers reporting 0 (accurate)

The issue here is the two extra time series, which should not exist if the nodepool doesn't exist.

4.12.35

This particular cluster had its OIDC configuration along with other customer AWS account resources deleted, which might be connected to the misbehaviour of the metric.

1.
2.
3.

Adding must-gather and metric time series in the ticket