Extend metrics for core resources

[bh-tt]

I'd like to extend the metrics for the builtin ResourceQuota (so we can make alerts if a namespace get close to the max and it starts to cause issues).

I tried to add these using the custom resources, yet I cannot enter an empty apiGroup in the groupVersionKind. That results in 'config.go:183] "failed to resolve GVK" err="group is required in the defined GVK /v1, Kind=ResourceQuota" gvk="_v1_ResourceQuota"'. I've also tried using group 'core', but that results in RBAC errors as the group really is "".

How can I add metrics like these to a standard resource? I suppose the alternative is sending a PR, but especially for this resource which can be easily extended with additional quotas for non-standard limits like GPUs that would be a significant maintenance burden on this project.

customresources config file:

    - apiVersion: v1
      kind: ConfigMap
      metadata:
        name: kube-state-crd-metrics
      data:
        crds.yaml: |
          spec:
            resources:
              - groupVersionKind:
                  group: ""
                  version: v1
                  kind: ResourceQuota
                labelsFromPath:
                  name: [metadata, name]
                  namespace: [metadata, namespace]
                metrics:
                  - name: quota_memory_limit
                    each:
                      type: Gauge
                      gauge:
                        path: [status,hard,"limits.memory"]
                  - name: quota_cpu_limit
                    each:
                      type: Gauge
                      gauge:
                        path: [status,hard,"limits.cpu"]
                  - name: quota_memory_request
                    each:
                      type: Gauge
                      gauge:
                        path: [status,hard,"requests.memory"]
                  - name: quota_cpu_request
                    each:
                      type: Gauge
                      gauge:
                        path: [status,hard,"requests.cpu"]
                  - name: used_memory_limit
                    each:
                      type: Gauge
                      gauge:
                        path: [status,used,"limits.memory"]
                  - name: used_cpu_limit
                    each:
                      type: Gauge
                      gauge:
                        path: [status,used,"limits.cpu"]
                  - name: used_memory_request
                    each:
                      type: Gauge
                      gauge:
                        path: [status,used,"requests.memory"]
                  - name: used_cpu_request
                    each:
                      type: Gauge
                      gauge:
                        path: [status,used,"requests.cpu"]

[robbat2]

I posted a PR to fix this:
#2435