Fix monitoring of K3s control plane components: kube-scheduler, kube-proxy, and kube-controller-manager

[matofeder]

PR#392 added Kubernetes monitoring deployment play. Kubernetes monitoring is intended to monitor the k3s instance control plane components out of the box.
PR#392 observes several issues that are mentioned in the PR's description, which are related to the k3s distribution.

This PR addresses the 1. issue where the monitoring solution could not monitor K3s control plane components kube-scheduler, kube-proxy, and kube-controller-manager. The second part of the fix has already been merged osism/defaults#177

Tesbed test

1. Update kubernetes-monitoring play

$ docker exec -it -u 0 osism-ansible bash
$ vim /ansible/infrastructure/kubernetes-monitoring.yml

2. Execute play

$ osism apply kubernetes-monitoring
2024-04-30 13:48:53 | INFO     | Task was prepared for execution.
2024-04-30 13:48:53 | INFO     | It takes a moment until the task has been started and output is visible here.

PLAY [Apply kubernetes monitoring role] ****************************************

TASK [Deploy kubernetes-monitoring helm chart] *********************************
Tuesday 30 April 2024  13:49:02 +0000 (0:00:03.935)       0:00:03.935 ********* 
changed: [localhost]

PLAY RECAP *********************************************************************
2024-04-30 13:50:26 | INFO     | Play has been completed. There may now be a delay until all logs have been written.
2024-04-30 13:50:26 | INFO     | Please wait and do not abort execution.
localhost                  : ok=1    changed=1    unreachable=0    failed=0    skipped=0    rescued=0    ignored=0   

Tuesday 30 April 2024  13:50:26 +0000 (0:01:23.529)       0:01:27.464 ********* 
=============================================================================== 
Deploy kubernetes-monitoring helm chart -------------------------------- 83.53s

3. Validate deployment

$ kubectl get po -n kubernetes-monitoring 
NAME                                                        READY   STATUS    RESTARTS   AGE
alertmanager-kube-prometheus-alertmanager-0                 2/2     Running   0          2m46s
dnation-kubernetes-jsonnet-translator-797fbf8ddb-msnqs      1/1     Running   0          2m49s
kube-prometheus-operator-85cbd9ffcd-dxmh5                   1/1     Running   0          2m49s
kubernetes-monitoring-grafana-566d9d4f65-gf2gw              3/3     Running   0          2m49s
kubernetes-monitoring-kube-state-metrics-79b6bfbc46-6mjcm   1/1     Running   0          2m49s
kubernetes-monitoring-prometheus-node-exporter-bdp9v        1/1     Running   0          2m49s
kubernetes-monitoring-prometheus-node-exporter-g4l4s        1/1     Running   0          2m49s
kubernetes-monitoring-prometheus-node-exporter-kgnb4        1/1     Running   0          2m49s
kubernetes-monitoring-prometheus-node-exporter-x5jc7        1/1     Running   0          2m49s
kubernetes-monitoring-thanos-query-6c979c7cc7-hhb6x         1/1     Running   0          2m49s
prometheus-kube-prometheus-prometheus-0                     3/3     Running   0          2m46s

4. Port forward Grafana UI, and access it (username: admin, password: pass)

$ export POD_NAME=$(kubectl get pods --namespace kubernetes-monitoring -l "app.kubernetes.io/name=grafana,app.kubernetes.io/instance=kubernetes-monitoring" -o jsonpath="{.items[0].metadata.name}")
$ kubectl --namespace kubernetes-monitoring port-forward --address 0.0.0.0 $POD_NAME 3000

See that all K3s control plane components kube-scheduler, kube-proxy, and kube-controller-manager are in OK state

Related to: SovereignCloudStack/issues#576