Revert #1166: un-mute the Advisor check now that #1165 is fixed

.github/workflows/synthetic-e2e.yml

@@ -62,16 +62,7 @@ jobs:
           echo "ADVISOR_PASSWORD=$adv" >> "$GITHUB_ENV"
           echo "MDR_PASSWORD=$mdr" >> "$GITHUB_ENV"
 
-      # #1165: the Advisor check still runs, but it no longer pages. Demo org1 learner
-      # data is duplicated thousands of times over (~6.3MB per lif_query), so every
-      # Advisor conversation exceeds the model context window and this check fails 100%
-      # of the time. A check that always fails is not a signal — it just trains people to
-      # filter the topic, which would also hide a real MDR/LDE outage. Its status is
-      # still reported in the run summary below.
-      # REMOVE continue-on-error THE MOMENT #1165 IS FIXED, or this check is dead weight.
       - name: Advisor happy path
-        id: advisor
-        continue-on-error: true
         env:
           BASE_URL: https://advisor.demo.lif.unicon.net/
           E2E_USERNAME: atsatrian_lifdemo@stateu.edu
@@ -86,19 +77,6 @@ jobs:
           MDR_PASSWORD: ${{ env.MDR_PASSWORD }}
         run: npx playwright test --grep @happy-path
 
-      # Keep the muted check visible. A mute that hides its own status is how a
-      # known issue quietly becomes an unknown one.
-      - name: Report Advisor check status (muted, #1165)
-        if: always()
-        working-directory: ${{ github.workspace }}
-        run: |
-          echo "### Advisor happy path: ${{ steps.advisor.outcome }}" >> "$GITHUB_STEP_SUMMARY"
-          if [ "${{ steps.advisor.outcome }}" = "failure" ]; then
-            echo "Muted under #1165 (demo org1 data duplication) — does not page." >> "$GITHUB_STEP_SUMMARY"
-          else
-            echo "Advisor is passing. If #1165 is fixed, remove continue-on-error from that step." >> "$GITHUB_STEP_SUMMARY"
-          fi
-
       - name: Upload Playwright report on failure
         if: failure()
         uses: actions/upload-artifact@v4
@@ -135,7 +113,7 @@ jobs:
           run_url="${GITHUB_SERVER_URL}/${GITHUB_REPOSITORY}/actions/runs/${GITHUB_RUN_ID}"
           aws sns publish --topic-arn "$topic" \
             --subject "LIF demo synthetic monitor FAILED (2+ consecutive)" \
-            --message "The demo MDR/LDE happy-path check has failed at least twice in a row. The Advisor check is muted under #1165 and is NOT covered by this alert. Investigate: ${run_url}"
+            --message "The demo happy-path monitor has failed at least twice in a row (advisor and/or MDR/LDE). Investigate: ${run_url}"
 
       # Send an "all clear" only when we actually alerted — i.e. the two previous
       # completed runs both failed, which is exactly the condition that fired the
@@ -156,12 +134,6 @@ jobs:
           fi
           topic=$(aws ssm get-parameter --name /demo/alerts/TopicArn --query Parameter.Value --output text)
           run_url="${GITHUB_SERVER_URL}/${GITHUB_REPOSITORY}/actions/runs/${GITHUB_RUN_ID}"
-          # Never claim a full recovery while the Advisor check is muted and still failing.
-          if [ "${{ steps.advisor.outcome }}" = "failure" ]; then
-            detail="The MDR/LDE happy-path check is passing again. The Advisor check is STILL FAILING but is muted under #1165, so it no longer pages."
-          else
-            detail="The demo happy-path monitor is passing again after a sustained failure."
-          fi
           aws sns publish --topic-arn "$topic" \
-            --subject "LIF demo synthetic monitor RECOVERED (MDR/LDE)" \
-            --message "${detail} ${run_url}"
+            --subject "LIF demo synthetic monitor RECOVERED" \
+            --message "The demo happy-path monitor is passing again after a sustained failure. ${run_url}"