Build/Test Tools: Restore automatically retrying failed E2E tests once.

This restores the GitHub Actions job responsible for automatically retrying a failed workflow once within the E2E testing workflow.

[56198] disabled Slack notifications for this workflow because of the increased number of timeout errors occurring after recent changes until they could be further investigated. Even though the signal-to-noise ration was way too high, there’s still benefit in retrying the workflow once to see if the timeout can be resolved without human intervention. The one retry attempt will not result in any Slack notifications.

Follow up to [56198].
See #58779.

git-svn-id: https://develop.svn.wordpress.org/trunk@56308 602fd350-edb4-49c9-b593-d223f7449a82

Author: desrosj

.github/workflows/end-to-end-tests.yml

@@ -126,3 +126,34 @@ jobs:
 
       - name: Ensure version-controlled files are not modified or deleted
         run: git diff --exit-code
+
+  failed-workflow:
+    name: Failed workflow tasks
+    runs-on: ubuntu-latest
+    permissions:
+      actions: write
+    needs: [ e2e-tests ]
+    if: |
+      always() &&
+      github.repository == 'WordPress/wordpress-develop' &&
+      github.event_name != 'pull_request' &&
+      github.run_attempt < 2 &&
+      (
+        needs.e2e-tests.result == 'cancelled' || needs.e2e-tests.result == 'failure'
+      )
+    steps:
+      - name: Dispatch workflow run
+        uses: actions/github-script@98814c53be79b1d30f795b907e553d8679345975 # v6.4.0
+        with:
+          retries: 2
+          retry-exempt-status-codes: 418
+          script: |
+            github.rest.actions.createWorkflowDispatch({
+              owner: context.repo.owner,
+              repo: context.repo.repo,
+              workflow_id: 'failed-workflow.yml',
+              ref: 'trunk',
+              inputs: {
+                run_id: '${{ github.run_id }}'
+              }
+            });