[quotamanagement] Handle case where tree has duplicate node names

[KPostOffice]

Currently when you have two nodes in a QuotaSubtree with the same name, such as:
tree.yaml

apiVersion: ibm.com/v1
kind: QuotaSubtree
metadata:
  name: context-root
  namespace: kube-system
  labels:
    tree: quota_context
spec:
  children:
    - name: context-root
      quotas:
        hardLimit: true
        requests:
          cpu: 2000
          memory: 8000Mi
---
apiVersion: ibm.com/v1
kind: QuotaSubtree
metadata:
  name: context-root-children
  namespace: kube-system
  labels:
    tree: quota_context
spec:
  parent: context-root
  children:
    - name: alpha
      quotas:
        hardLimit: true
        requests:
          cpu: 100
          memory: 4000Mi
    - name: beta
      quotas:
        hardLimit: true
        requests:
          cpu: 100
          memory: 4000Mi

---

node-dupes.yaml

apiVersion: ibm.com/v1
kind: QuotaSubtree
metadata:
  name: context-root-children-dupes
  namespace: kube-system
  labels:
    tree: quota_context
spec:
  parent: context-root
  children:
    - name: alpha
      quotas:
        hardLimit: true
        requests:
          cpu: 1000
          memory: 4000Mi
    - name: beta
      quotas:
        hardLimit: true
        requests:
          cpu: 1000
          memory: 4000Mi

they act as though both quotas are applied within an AppWrapper effectively making it the strictest subset of the two. This behavior should either be accepted and documented clearly , or the qst should be rejected and be given a clear condition as to why.

[asm582]

@metalcycling Can you please comment here based on your experience with quota management?