Add some examples of per-queue alerts/dashboards

[binarin]

Summary Of Changes

Add examples of how to use per-queue metrics introduced in rabbitmq/rabbitmq-server#3421

This includes:

1. Configuring new prometheus scrape targes
2. A few alert using new metrics
3. A grafana dashboard where a single queue can be observed

[yaronp68]

one suggestion:
In the Grafana dashboard dropdown, when I select a RabbitMQ Cluster, I can still see queues defined in other RabbitMQ clusters.
I’d be nice to filter queues by my selected RabbitmqCluster.
If you edit the variable query for queue which is currently:
query_result(rabbitmq_detailed_queue_messages{namespace="$namespace"} * on (instance, job) group_left(rabbitmq_cluster) rabbitmq_identity_info{namespace="$namespace"})
to
query_result(rabbitmq_detailed_queue_messages{namespace="$namespace"} * on (instance, job) group_left(rabbitmq_cluster) rabbitmq_identity_info{namespace="$namespace", rabbitmq_cluster="$rabbitmq_cluster"})
that should work.
similarly as we list only rabbitmqclusters in the dropdown for the selected namespace