Show which experiment each run used on the 5000-node experimental scale job in TestGrid

The `gce-master-scale-performance-5000-experimental` job is our exploratory scale job, and we change its configuration every few weeks (recent ones: apiserver restart, range streaming, baseline). Because it is one job that we keep reconfiguring, TestGrid shows a single undifferentiated series and there is no way to tell which experiment produced a given run. https://testgrid.k8s.io/sig-scalability-gce#gce-master-scale-performance-5000-experimental <img width="1026" height="316" alt="Image" src="https://github.com/user-attachments/assets/d56aa335-bd51-4122-af9b-433257d6cb7a" /> We want each run labeled with the experiment it was running, shown as a new column header above that run's column in TestGrid, so we can read the configuration history at a glance without splitting the job. This builds on the same per-run metadata reporting that `ci-kind-dra` already does (https://github.com/kubernetes/test-infra/blob/master/config/jobs/kubernetes/sig-node/dra-ci.yaml#L35), plus a new column header to display it. Two small changes: - kubernetes/kops: have the shared scalability `run-test.sh` report the experiment name from an env var into the run metadata. - kubernetes/test-infra: set that env var on the experimental job and declare the new column header on its TestGrid config.