Provide a mechanism for operands to get control plane affinity and priority rules

Context:

https://coreos.slack.com/archives/G01QS0P2F6W/p1661437729197269

We have no means to apply our affinity and priority opinions to operands running in the management cluster 

AC:

All operands have the appropriate affinity and priority rules. - Discussed 27 Feb this might be a webook.

Notes from https://issues.redhat.com/browse/HOSTEDCP-909

I am testing moving an HCP to a new machinepools in Hypershift as outlined here (https://docs.google.com/document/d/1tNvoNInor9lNswqGVBN13z2XKyjI4w4zumk7pKJRzJs/edit#). During this testing I found that the following deployments and statefulset did not have the expected  toleration of

hypershift.openshift.io/control-plane: true

and

hypershift.openshift.io/cluster: ${HostedControlPlane Namespace}

as defined here (https://hypershift-docs.netlify.app/how-to/distribute-hosted-cluster-workloads/)

Deployments missing toleration:

audit-webhook

aws-ebs-csi-driver-controller

aws-ebs-csi-driver-operator

cloud-network-config-controller

csi-snapshot-controller

csi-snapshot-webhook

metrics-forwarder-deployment

multus-admission-controller

Statefulsets missing toleration:

ovnkube-master

These tolerations are required for successful isolation of HCPs on Hypershift.