[kube-prometheus-stack] etcd Multi-cluster dashboard

[dragoangel]

Is your feature request related to a problem ?

At the moment of writing buildin dasbhoard do utilize {job=~".*etcd.*", job="$cluster"} even cluster variable already filtered globally by .*etcd.* and is actually a job, while cluster as name of variable need to be used to differentiate real clusters, as result I have 2 k8s clusters and all etcds are joined by one name of job.

Describe the solution you'd like.

Adjust etcd dashboard to utilize $cluster and $job properly.

Describe alternatives you've considered.

N/A

Additional context.

I have multicluster enabled in grafana sidecar settings for both global and etcd.