Add k8s settings: prefer-closest-numa-nodes and max-allowable-numa-nodes

[piyush-jena]

Issue number:

Closes # bottlerocket-os/bottlerocket#4750

Description of changes:
Add 2 topology manager policy options:

1. max-allowable-numa-nodes - GA k8s-1.35+
2. prefer-closest-numa-nodes - GA k8s-1.32+

Testing done:
Enabled the following settings and rebooted the instance:

apiclient set \
  kubernetes.cpu-manager-policy="static" \
  kubernetes.topology-manager-policy="best-effort" \
  kubernetes.topology-manager-policy-options.prefer-closest-numa-nodes="true"

Then ran a pod with the following spec:

apiVersion: v1
kind: Pod
metadata:
  name: numa-test
spec:
  containers:
  - name: numa-test
    image: registry.k8s.io/pause:3.9
    resources:
      requests:
        cpu: "4"
        memory: "100Mi"
      limits:
        cpu: "4"
        memory: "100Mi"

Its important that requests and limits are equal for both CPU and memory to ensure it is in the Guaranteed QoS class for the CPU pinning as a result of the static policy to work properly.

The assigned CPUs for this pod was 1,2,65,66.
The NUMA node ranges for this instances are:
node0: 0-31, 64-95
node1: 32-63, 96-127

All 4 CPUs belong to NUMA node0. The prefer-closest-numa-nodes policy worked as expected.

For max-allowable-numa-nodes I show the limits enforced by kubernetes are also applied here.

[ssm-user@control]$ apiclient set settings.kubernetes.topology-manager-policy-options.max-allowable-numa-nodes="4"
Failed to change settings: Failed PATCH request to '/settings/keypair?tx=apiclient-set-Z6sWPcf69TAMvS9h': Status 400 when PATCHing /settings/keypair?tx=apiclient-set-Z6sWPcf69TAMvS9h: Unable to match your input to the data model.  We may not have enough type information.  Please try the --json input form.  Cause: Error during deserialization: Invalid Kubernetes max-allowable-numa-nodes value '4': must be >= 8 at line 1 column 79
[ssm-user@control]$ apiclient set settings.kubernetes.topology-manager-policy-options.max-allowable-numa-nodes="8"

Terms of contribution:

By submitting this pull request, I agree that this contribution is dual-licensed under the terms of both the Apache License, version 2.0, and the MIT license.