Cluster Logging Operator should be coupled loosely with Cluster Monitoring Prometheus in "openshift-logging" on ROSA

Describe the bug
Cluster Logging Operator try to add alert rules to configmap in "openshift-monitoring" regardless of running Platforms.
For example, even though Cluster Logging Operator(EFK stack) on ROSA is not in SRE responsibility, so the alert rules should be added only to the configmap in "openshift-user-workload-monitoring".

Actually, if OpenShift Logging is installed on ROSA, the following error messages spam is keep to record into Cluster Logging Operator pod logs.

{"component":"cluster-logging-operator","error":{"msg":"Unable to create or update collection for \"instance\": failure creating the fluentd PrometheusRule: Internal error occurred: failed calling webhook \"prometheusrules.openshift.io\": Post \"https://prometheus-operator.openshift-monitoring.svc:8080/
admission-prometheusrules/validate?timeout=5s\": no endpoints available for service \"prometheus-operator\""},"level":"0","message":"Error reconciling clusterlogging instance","ts":"2021-07-02T02:33:12.553198443Z"}

Environment

• Red Hat OpenShift Service on AWS (ROSA) , 4.7.z
• OpenShift Logging Operator (5.0)

Expected behavior
No spam error messages, and Cluster Logging Operator can specify the target configmap using an option.

Actual behavior
Cluster Logging Operator try to add the alert rules endlessly into the configmap in "openshift-monitoring" through "https://prometheus-operator.openshift-monitoring.svc:8080/". And it's built-in. There is no option to change the target prometheus service at the moment.

To Reproduce
You can always reproduce this issue as deploying OpenShift Logging Operator at ROSA.

Additional context
This issue is also related with the following ticket.

1. LOG-1561