Feat/granular dashboards

[puffitos]

Motivation

Implements #25

Not all dashboards are always needed; for example, a cluster may not have longhorn installed or the performance dashboards make no sense in a downstream cluster, as noted in the issue.

Changes

• Split up the configmap dashboard file into multiple files
• Added value toggles to activate/deactivate dashboards

Tests done

All active

Deployed a version with the standard values, all dashboards are present again:

I'll do another test with the performance dashboards deactivated (and all others as well) to see how helm and our grafana reacts.

Rancher Performance deactivated

Changes the values to:

caas:
  grafana:
    dashboards:
      rancherPerformance: false

The configmap was removed and grafana sucessfully refreshed the dashbhoards. No more rancher perfomance:

All rancher dashboards gone

Updated the values to:

caas:
  grafana:
    dashboards:
      nginxIngress: false
      rancherCluster: false
      rancherHome: false
      rancherK8sComponents: false
      rancherNodes: false
      rancherPerformance: false
      rancherPods: false
      rancherWorkloads: false
      longhorn: false

and, as expected, the dashboards were also removed from grafana:

[puffitos]

BTW, this value wasn't used anymore 🤷