Add Etcd client support: stop traffic to in-defragmentation-server

[Yiqing-Liu]

What would you like to be added?

Etcd defragmentation is the workflow to prevent etcd from running out of disk space.
It is expected that defragmentation will result in timeouts and latency to requests that happen to be connected to the node being defragmented.

There is a proposal to stop traffic sending to being-defragmented-etcd-node.
etcd-io/etcd#16278

The Etcd server's implementation has been merged in etcd-io/etcd#17914, which is guarded by a feature flag.

Propose client side make the corresponding change (which can be guarded by feature flag as well) to complete the implementation.

Why is this needed?

This work will largely mitigate the Etcd defrag impact, decreasing the 5xx error rate.
to address the reported issue: #93280

[k8s-ci-robot]

This issue is currently awaiting triage.

If a SIG or subproject determines this is a relevant issue, they will accept it by applying the triage/accepted label and provide further guidance.

The triage/accepted label can be added by org members by writing /triage accepted in a comment.

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes-sigs/prow repository.

[pranav-pandey0804]

/sig api-machinery