We need to always delete the kind cluster if we create kind cluster with Prow jobs

Currently when we run test flows with `kind`, we run `kubetest2 kind --up --down` which will also delete the kind cluster after the test flow finishes in the normal way. But if the Prow job is interrupted in the middle, the cluster will not be properly deleted thus will be causing memory leakage, as described in https://github.com/kubernetes-sigs/kind/issues/303#issuecomment-521384993

We should add a trap command to make sure the cluster can still be cleaned up even when the Prow job is interrupted.

Start by locating the Prow test-flow entry point that runs `kubetest2 kind --up --down`, then read how the normal cleanup is performed. Check the referenced kind interruption behavior and add a cleanup trap at the appropriate script boundary. Done means an interrupted Prow job still deletes the kind cluster, while normal test flows continue to clean up.

Written by the indexing model from the issue text.