ztunnel workload detail metrics

[josunect]

Describe the change

Initial set of ztunnel metrics:

Added grafana link when the ztunnel dashboard is included:

  grafana:
    dashboards:
     ...
      - name: "Istio Ztunnel Dashboard"

Steps to test the PR

(Note: A kind cluster with more than one node can be created for more metrics):

kind create cluster --config=- <<EOF
kind: Cluster
apiVersion: kind.x-k8s.io/v1alpha4
name: ambient
nodes:
- role: control-plane
- role: worker
- role: worker
EOF

• Install Istio Ambient
• Install Kiali
• If Grafana is installed, include the ztunnel dashboard in the CR:

  grafana:
    dashboards:
     ...
      - name: "Istio Ztunnel Dashboard"

• Go to workload ztunnel in Istio namespace, click in ztunnel tab, click in Metrics subtab.

Automation testing

If applicable, explain the case scencarios covered by unit / integration / e2e tests created for this PR.

Issue reference

Fixes #8145

[josunect]

There is something that I don't have clear how to do: If is possible to add a label in the by {pod} query. This is because, in the bytes transmitted, it includes the tcp sent bytes and the tcp received (So there are double data with the same label, which is the pod name) Updated with a metric conversion label prefix

Besides that, I think it is ready for a first review, at least to have some feedback about the content.
It can be tested with kind so there are more than one ztunnel and it has more rich data:

kind create cluster --config=- <<EOF
kind: Cluster
apiVersion: kind.x-k8s.io/v1alpha4
name: ambient
nodes:
- role: control-plane
- role: worker
- role: worker
EOF

[hhovsepy]

looks good,

[hhovsepy]

Verified.
Custom time range works fine.

[jshaughn]

Is it my imagination or are the charts squished vertically? Memory and CPU are not showing up for me, but maybe that's true on minikube. Although, I do have those for my istiod.

[hhovsepy]

Still works fine on Kind cluster:

Will wait for a review from @jshaughn

[jshaughn]

Now seeing the cpu/metrics charts on mkube, so that's good. And the y-axis on the charts looks good. The charts are larger than the other metrics charts, but I think that's OK given that we only jave 6.

The problem I'm seeing is that the full-screen expansion is too big:

Finally, I wonder whether we should disable the other traffic and metric tabs for ztunnel, will they ever have data?

[josunect]

Now seeing the cpu/metrics charts on mkube, so that's good. And the y-axis on the charts looks good. The charts are larger than the other metrics charts, but I think that's OK given that we only jave 6. The problem I'm seeing is that the full-screen expansion is too big: Finally, I wonder whether we should disable the other traffic and metric tabs for ztunnel, will they ever have data?

Thanks, @jshaughn , I have updated the styles also for the scroll.
Regarding hidden the traffic and metric tabs, should this be done also for istiod? Or even for all the istio-system namespace?