feat: adding kube prometheus monitoring stack

[lmyslinski]

This PR introduces a new monitoring setup, which primarily replaces our existing infrastructure setup. Instead of deploying everything via Pulumi, we utilize a different tool for setting up this part of the infrastructure configuration with a dedicated tool called kube-prometheus.

This middle-ground solution allows us to tweak most of the settings more easily and use built-in configuration options. For things that are not covered by the tool, we still depend on the Pulumi stack, such as the Traeffic ingress or any additional stuff that's easier to do via Pulumi.

Changes performed:

• New tool added to the unchained monitoring setup
• CI configured to apply the changes to our cluster after approval
• Because kube-prometheus does not offer diff functionality similar to Pulumi, we depend on kubectl diff to preview the changes to be made to our stack
• Removal of the old tool setup

Closes #700

[kaladinlight]

to clarify, does the kube prometheus actually deploy grafana/loki/prometheus, or just bundle grafana dashboard views and different alerts?

[lmyslinski]

to clarify, does the kube prometheus actually deploy grafana/loki/prometheus, or just bundle grafana dashboard views and different alerts?

It's the full stack. Prometheus operator, node exporter, grafana, alert manager plus some nice default dashboards and alerts