Currently we check for the version number to determine whether we should do automatic restarts for certificate expiration.
https://github.com/openshift/hypershift/blob/6991b4a6bb8ead41e1e5a0843975b7761142514f/hypershift-operator/controllers/hostedcluster/hostedcluster_controller.go#L1978
This should instead use a CPO image label check that is present whenever we move to use CPO v2 

    4.19

    Always

    1. Create a HostedCluster test cluster with 4.19 nightly and annotations to shorten certificate expiration.
    2. Wait for clusters to expire.
    3.
    

    No remediation occurs when certs expire

    Remediation occurs and cluster continues to function properly