docs(kernel-e2e): refresh workflow header — setup is done, queue is live

The "Required external setup" block read like a TODO list, but every
item on it is in place after #808 landed and the main ruleset was
updated with required_status_checks + merge_queue. Reword as a
"things this depends on (debugging crib)" and add the ruleset note
so future readers know why the merge_group path matters.

Also serves as a low-risk pilot PR for exercising the new merge
queue end-to-end — comments-only diff, no behavioral change.

Co-authored-by: Isaac
Signed-off-by: Vikrant Puppala <vikrant.puppala@databricks.com>

Author: vikrantpuppala

.github/workflows/kernel-e2e.yml

@@ -20,18 +20,17 @@ name: Kernel E2E Tests
 #   - merge_group fires the real gate — dispatches when kernel-relevant
 #     files changed, auto-passes otherwise.
 #
-# Required external setup:
-#   1. `kernel-e2e` label exists in this repo.
-#   2. `INTEGRATION_TEST_APP_ID` / `INTEGRATION_TEST_PRIVATE_KEY`
-#      secrets exist (already installed for the proxy-tests workflow).
-#      The GitHub App's repo allowlist must include
-#      `databricks/databricks-sql-kernel` — extend the existing App
-#      config; do not create a new App.
-#   3. `KERNEL_REV` file at the repo root containing a 40-char kernel
-#      commit SHA.
-#   4. `azure-prod` environment exposes DATABRICKS_HOST /
-#      TEST_PECO_WAREHOUSE_HTTP_PATH / DATABRICKS_TOKEN
-#      (already configured for code-coverage.yml).
+# External setup that this workflow depends on (already in place,
+# documented here for future debugging):
+#   - `kernel-e2e` label exists in this repo.
+#   - `INTEGRATION_TEST_APP_ID` / `INTEGRATION_TEST_PRIVATE_KEY`
+#     secrets are installed and the App's repo allowlist includes
+#     `databricks/databricks-sql-kernel`.
+#   - `KERNEL_REV` file at the repo root pins the kernel commit SHA.
+#   - `azure-prod` environment exposes DATABRICKS_HOST /
+#     TEST_PECO_WAREHOUSE_HTTP_PATH / DATABRICKS_TOKEN.
+#   - `Kernel E2E` is listed as a required status check on the
+#     `main` ruleset, so merge queue waits for it.
 
 on:
   pull_request: