Added label to all-pods service to narrow metrics scrape selection

[itzg]

This change adds a label cassandra.datastax.com/prom-metrics: true to the {cluster}-{dc}-all-pods-service in order to narrow the service selector in the Prometheus operator's ServiceMonitorSpec. Prior to this change both of the services in the following listing would get selected and scraped:

NAME                            TYPE        CLUSTER-IP     EXTERNAL-IP   PORT(S)                      AGE   LABELS
cluster1-dc1-all-pods-service   ClusterIP   None           <none>        9042/TCP,8080/TCP,9103/TCP   90m   app.kubernetes.io/managed-by=cass-operator,cassandra.datastax.com/cluster=cluster1,cassandra.datastax.com/datacenter=newflood-dc1
cluster1-dc1-service            ClusterIP   None           <none>        9042/TCP,8080/TCP,9103/TCP   90m   app.kubernetes.io/managed-by=cass-operator,cassandra.datastax.com/cluster=cluster1,cassandra.datastax.com/datacenter=newflood-dc1

cc @jimdickinson

Fixes #276

[itzg]

One of the check failures seems to be beyond my control

https://github.com/datastax/cass-operator/runs/1228413780?check_suite_focus=true#step:10:9

[jimdickinson]

@itzg thank you! could you sign the DataStax CLA at https://cla.datastax.com ?

[itzg]

@jimdickinson I have now signed the CLA.

[jimdickinson]

One of the check failures seems to be beyond my control

https://github.com/datastax/cass-operator/runs/1228413780?check_suite_focus=true#step:10:9

Sorry I dragged my feet on this one - I looked into this issue and there was nothing to do about it. We'll re run tests on master and make sure all is well. Thanks for the contribution!