Propagate log levels from k8s dependencies

[der-eismann]

Describe the bug
When looking at the logs I can see that all messages directly from NTH have a level like INF or WRN. However if the message is coming from a k8s dependency (I think), the log level is given with ??? or omitted for JSON logging. It would be great if we could have a log level for these messages as well.

Steps to reproduce

1. Run NTH and drain a node

Expected outcome
There should be no ??? for log levels anymore

Application Logs

2025/03/25 08:07:27 INF Draining the node
2025/03/25 08:07:27 ??? WARNING: ignoring DaemonSet-managed Pods: default/abc-wlm7t, default/def-rl86f, default/ghi-bb65s, …
2025/03/25 08:07:27 ??? evicting pod default/my-pod-79dc9db6d5-g9rln

Environment

• NTH App Version: v1.25.0
• NTH Mode (IMDS/Queue processor): Queue
• OS/Arch: Container / Bottlerocket arm64
• Kubernetes version: 1.32
• Installation method: Helm

[der-eismann]

Solution may be the same as for #1150 with the SetLogger function in klog

[der-eismann]

Turns out that this doesn't help as this comes from kubectl and that's using plain fmt and not klog 😕