fix: use latest Tekton releases and fix ConfigMap timing issue

Update E2E test setup to use latest stable releases for both Triggers and Pipelines, and fix the ConfigMap timing issue that causes triggers controller to crash:

**Pipelines:**
- Remove v1.12.0 pin and use latest release
- Remove TODO comment about switching back to latest

**Triggers:**
- Skip nightly builds and use latest release directly
- Fix ConfigMap timing: apply interceptors.yaml BEFORE release.yaml

**Root Cause:**
The triggers controller was crashing with: `configmap "config-triggers-core-interceptors" not found`

The ConfigMap is in interceptors.yaml, but when we applied release.yaml first, Kubernetes immediately started the controller pods. They crashed before interceptors.yaml could be applied.

**Fix:**
1. Apply interceptors.yaml first (creates ConfigMap/Secret; ClusterInterceptor resources fail without CRD)
2. Apply release.yaml (controller now finds ConfigMap and starts successfully)
3. Wait for ClusterInterceptor CRD to be established
4. Apply interceptors.yaml again (ClusterInterceptor resources now succeed)

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>

Author: divyansh42

test/e2e-common.sh

@@ -134,11 +134,8 @@ function install_pipeline_crd() {
   if [[ -n ${RELEASE_YAML_PIPELINE} ]];then
 	latestreleaseyaml=${RELEASE_YAML_PIPELINE}
   else
-
-    # Pin to v1.12.0 to avoid spurious termination message WARN in step logs (tektoncd/pipeline#10159).
-    # The fix (tektoncd/pipeline#10160) is merged but not yet in a released version.
-    # TODO: switch back to latest once a pipeline release includes the fix.
-    latestreleaseyaml=https://infra.tekton.dev/tekton-releases/pipeline/previous/v1.12.0/release.yaml
+    # Use latest release version
+    latestreleaseyaml="https://infra.tekton.dev/tekton-releases/pipeline/latest/release.yaml"
   fi
   [[ -z ${latestreleaseyaml} ]] && fail_test "Could not get latest released release.yaml"
   kubectl apply -f ${latestreleaseyaml} ||
@@ -160,34 +157,36 @@ function install_triggers_crd() {
   if [[ -n ${RELEASE_YAML_TRIGGERS} ]];then
 	latestreleaseyaml=${RELEASE_YAML_TRIGGERS}
   else
-    # First try to install latestreleaseyaml from nightly
-    curl -o/dev/null -s -LI -f https://infra.tekton.dev/tekton-releases-nightly/triggers/latest/release.yaml &&
-        latestreleaseyaml=https://infra.tekton.dev/tekton-releases-nightly/triggers/latest/release.yaml
-
-    # If for whatever reason the nightly release wasnt there (nightly ci failure?), try the released version
-    [[ -z ${latestreleaseyaml} ]] && latestreleaseyaml="https://infra.tekton.dev/tekton-releases/triggers/latest/release.yaml"
+    # Use latest release version (not nightly)
+    latestreleaseyaml="https://infra.tekton.dev/tekton-releases/triggers/latest/release.yaml"
   fi
   if [[ -n ${RELEASE_YAML_TRIGGERS_INTERCEPTORS} ]];then
 	latestinterceptorsyaml=${RELEASE_YAML_TRIGGERS_INTERCEPTORS}
   else
-    # First try to install latest interceptors from nightly
-    curl -o/dev/null -s -LI -f https://infra.tekton.dev/tekton-releases-nightly/triggers/latest/interceptors.yaml &&
-        latestinterceptorsyaml=https://infra.tekton.dev/tekton-releases-nightly/triggers/latest/interceptors.yaml
-
-    # If for whatever reason the nightly release wasnt there (nightly ci failure?), try the released version
-    [[ -z ${latestinterceptorsyaml} ]] && latestinterceptorsyaml="https://infra.tekton.dev/tekton-releases/triggers/latest/interceptors.yaml"
+    # Use latest release version (not nightly)
+    latestinterceptorsyaml="https://infra.tekton.dev/tekton-releases/triggers/latest/interceptors.yaml"
   fi
   [[ -z ${latestreleaseyaml} ]] && fail_test "Could not get latest released release.yaml"
   [[ -z ${latestinterceptorsyaml} ]] && fail_test "Could not get latest released interceptors.yaml"
+
+  # Apply interceptors.yaml first to create ConfigMap that triggers controller needs
+  # ClusterInterceptor resources will fail (CRD doesn't exist yet) but ConfigMap/Secret will be created
+  kubectl apply -f ${latestinterceptorsyaml} 2>/dev/null || true
+
+  # Apply release.yaml to install CRDs and controllers
+  # Controller will now find the ConfigMap and start successfully
   kubectl apply -f ${latestreleaseyaml} ||
     fail_test "Build triggers installation failed"
 
-  # Wait for pods to be running in the namespaces we are deploying to
-  wait_until_pods_running tekton-pipelines || fail_test "Tekton Triggers did not come up"
-
+  # Wait for CRD to be established
   kubectl wait --for=condition=Established --timeout=30s crds/clusterinterceptors.triggers.tekton.dev || fail_test "cluster interceptors never established"
+
+  # Apply interceptors.yaml again to create ClusterInterceptor resources (now that CRD exists)
   kubectl apply -f ${latestinterceptorsyaml} || fail_test "Interceptors installation failed"
 
+  # Wait for pods to be running in the namespaces we are deploying to
+  wait_until_pods_running tekton-pipelines || fail_test "Tekton Triggers did not come up"
+
   # Make sure that eveything is cleaned up in the current namespace.
   for res in eventlistener triggertemplate triggerbinding clustertriggerbinding; do
     kubectl delete --ignore-not-found=true ${res}.triggers.tekton.dev --all