Alloy-Mixin: allow k8s cluster and alloy cluster disable, add logs dashboard

[gaantunes]

PR Description

Introducing the following changes:

• including a config file to the mixin, which allows you to:
  • disable dashboards, panels and alerts that are targeted on monitoring alloy instances being run in cluster mode.
  • disable the inclusion of k8s cluster targeted labels on dashboards and alerts.
  • change the dashboard tags
• including a logs dashboard
• including a job label everywhere as a grouper when cluster and namespace are disabled, to avoid metric conflicts - like up metric

Which issue(s) this PR fixes

Notes to the Reviewer

Let me know if this needs changelog updates or if there are any docs.

PR Checklist

• CHANGELOG.md updated
• Documentation added
• Tests updated
• Config converters updated

[rfratto]

Thanks! This is a bit hard to review for concrete differences at the moment given how many things changed, but I left a few comments to help reduce the number of differences so I can give a more thoughtful second pass.

[rfratto]

Thanks for the changes!

[rfratto]

This doesn't seem to follow the pattern that other dashboards do; this is the only one with a grafanaDashboards top-level key. Can we change this to be consistent with the other dashboards?

That would also mean moving the logic for whether to enable the dashboard outside of the definition of the dashboard itself, which would be more consistent with how the clustering dashboards are made optional.

[gaantunes]

That is because the library I am using expects a grafanaDashboards variable.
I also don't like how it looks different, but since this is a totally different library which is synced with the latest version of grafonnet, I believe it is an acceptable trade off.

[rfratto]

No other comments from me, nice work making this configurable!