flaky test in prefix plugin

[nirrozenbaum]

What happened:
flaky test in prefix plugin - I made PR #1528 which only updates few log lines and prefix unit test failed.
more information is here:
https://prow.k8s.io/view/gs/kubernetes-ci-logs/pr-logs/pull/kubernetes-sigs_gateway-api-inference-extension/1528/pull-gateway-api-inference-extension-test-unit-main/1963475514790252544

What you expected to happen:
tests should always pass
How to reproduce it (as minimally and precisely as possible):

Anything else we need to know?:

Environment:

• Kubernetes version (use kubectl version):
• Inference extension version (use git describe --tags --dirty --always):
• Cloud provider or hardware configuration:
• Install tools:
• Others:

[nirrozenbaum]

cc: @liu-cong @kfswain @kfirtoledo

[nirrozenbaum]

another link to PR #1524 (which only updates quickstart guide) with this flake :
https://prow.k8s.io/view/gs/kubernetes-ci-logs/pr-logs/pull/kubernetes-sigs_gateway-api-inference-extension/1524/pull-gateway-api-inference-extension-test-unit-main/1963489858903085056

[nirrozenbaum]

from my short analysis -
I think the introduction of the go routine in prefix plugin introduces race conditions in unit-tests

go func() {
		p.indexer.Add(state.PrefixHashes, ServerID(targetPod.NamespacedName))
	}()

the prefix plugin isn't finishing to update the indexer and the test assumes the state is already saved.