jobs: avoid blocking job which doesn't always run

As bentheelder said on
Slack (https://kubernetes.slack.com/archives/C2C40FMNF/p1734418617113169?thread_ts=1734417601.687079&cid=C2C40FMNF),
"adding blocking tests that only sometimes run is confusing and doesn't work
well".

That's because a regression might get merged without running the job. Then
any PR where the job gets run is blocked, whether it is related to the failure
or not.

Therefore pull-kubernetes-e2e-gce-network-proxy-http-connect should not be
blocking.

Author: pohly

config/jobs/kubernetes/sig-api-machinery/sig-api-machinery-config.yaml

@@ -80,6 +80,7 @@ presubmits:
     cluster: k8s-infra-prow-build
     always_run: false
     run_if_changed: '^(cluster/gce/manifests/konnectivity-server.yaml$|cluster/gce/addons/konnectivity-agent)'
+    optional: true # Doesn't always run and regressions shouldn't block merging PRs where it gets run.
     labels:
       preset-service-account: "true"
       preset-k8s-ssh: "true"

config/testgrids/kubernetes/presubmits/config.yaml

@@ -48,9 +48,6 @@ dashboards:
   - name: pull-kubernetes-dependencies
     test_group_name: pull-kubernetes-dependencies
     base_options: width=10
-  - name: pull-kubernetes-e2e-gce-network-proxy-http-connect
-    test_group_name: pull-kubernetes-e2e-gce-network-proxy-http-connect
-    base_options: width=10
   - name: pull-kubernetes-conformance-kind-ga-only-parallel
     test_group_name: pull-kubernetes-conformance-kind-ga-only-parallel
     base_options: width=10