#0: [skip ci] Add initial auto-retry post commit workflow

.github/workflows/_auto-retry-post-commit.yaml

@@ -0,0 +1,72 @@
+name: "[internal] Auto-retry post-commit"
+
+on:
+  workflow_dispatch:
+    inputs:
+      test_workflow_run_id:
+        description: "Unique GitHub workflow run ID to use for test"
+        default: 1278872273u
+        type: number
+      test_workflow_run_attempt:
+        description: "Run attempt of the workflow run"
+        default: 1
+        type: number
+  workflow_run:
+    workflows:
+      - "All post-commit tests"
+    types:
+      - completed
+
+jobs:
+  auto-retry:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+      - name: Get workflow run_id attempt number to analyze
+        id: get-run-id-and-attempt
+        shell: bash
+        run: |
+          event_name="${{ github.event_name }}"
+          if [[ "$event_name" == "workflow_dispatch" ]]; then
+            run_id="${{ inputs.test_workflow_run_id }}"
+            attempt_number="${{ inputs.test_workflow_run_attempt }}"
+          elif [[ "$event_name" == "workflow_run" ]]; then
+            run_id="${{ github.event.workflow_run.id }}"
+            attempt_number="${{ github.event.workflow_run.run_attempt }}"
+            [[ -z "$run_id" ]] && { echo "run_id is empty" ; exit 1; }
+            [[ -z "$attempt_number" ]] && { echo "attempt_number is empty" ; exit 1; }
+          else
+            echo "Unknown event name" && exit 1
+          fi
+
+          echo $run_id
+          echo $attempt_number
+          echo "run-id=$run_id" >> "$GITHUB_OUTPUT"
+          echo "attempt-number=$attempt_number" >> "$GITHUB_OUTPUT"
+
+          echo "::notice title=target-workflow-link::The workflow being analyzed is available at https://github.com/tenstorrent/tt-metal/actions/runs/$run_id/attempts/$attempt_number"
+      - name: Determine if we should continue
+        shell: bash
+        run: |
+          MAX_ATTEMPTS=5
+          if [[ "${{ steps.get-run-id-and-attempt.outputs.attempt_number }}" == "$MAX_ATTEMPTS" ]]; then
+            echo "::notice title=no-continue-max-tries::This workflow has exceeded max tries. Not re-trying"
+            should_continue=false
+          elif [[ "${{ github.event_name }}" != "workflow_dispatch" ]]; then
+            echo "::notice title=no-continue-is-on-branch::This workflow was not dispatched - not retrying"
+            should_continue=false
+          else
+            should_continue=true
+          fi
+
+          echo "should-continue=$should_continue" >> "$GITHUB_OUTPUT"
+      - name: Re-run failed jobs
+        env:
+          GH_TOKEN: ${{ github.token }}
+        run: |
+          echo "Re-running jobs here"
+      - uses: ./.github/actions/slack-report
+        if: ${{ failure() }}
+        with:
+          slack_webhook_url: ${{ secrets.SLACK_WEBHOOK_URL }}
+          owner: U014XCQ9CF8 # tt-rkim