As a cluster admin, I would like Cluster Resource Override to run as a deployment instead of a daemonset. That way the number of running replicas isn't necessarily tied to the number of masters in the cluster.

Acceptance Criteria

The new deployment should be running with 2 replicas on masters (by default – another story will cover the ability to control where else it can run).

The pods should have an anti-affinity which prevents them from running on the same host, like the example below:

template:
  spec:
    affinity:
      podAntiAffinity:
        requiredDuringSchedulingIgnoredDuringExecution:
        - labelSelector:
            matchExpressions:
            - key: clusterresourceoverride
              operator: In
              values:
              - "true"
          topologyKey: kubernetes.io/hostname