Bring your own Prometheus

Come up with a working configuration of OSSM with an external prometheus instance. Grafana is out of scope for this story: we assume that the user has a working grafana instance connected to their prometheus.

Take into account that if mTLS is set to STRICT in a mesh, we currently have to add a sidecar to prometheus to enable it to talk to the services. We might be able to work around that fact by using the prometheus merge feature of istio-proxy, where it serves metrics unencrypted.

Acceptance Criteria:

• Application & mesh metrics are present in prometheus
• Kiali retrieves its metrics from the external prometheus instance (optional)
• Integration or Acceptance test that verifies it
• The solution works with the default Cluster Monitoring prometheus instance
• create an OSSMDOCS issue that contains all relevant information for the docs (should probably go into the 'Deploying to Production' section)
  • Note: Docs issue already created for this epic: OSSMDOC-484

For QE:
All steps to verify integration with OpenShift Monitoring or custom Prometheus Operators are prepared here: https://github.com/maistra/istio-operator/pull/1094.