fix(ci): retry Pages deployment once on transient failure

The Pages backend intermittently rejects deployments with a generic
'Deployment failed, try again later'. Both workflow_run-triggered
deploys so far (the post-release appcast publishes for v1.16.1 and
v1.17.0) hit this, leaving the live appcast stale so Sparkle stopped
offering the new release until an unrelated push redeployed the site.
A manual rerun of the identical job succeeds, so retry once in-job
after a 60 second pause before failing the run.

Author: bocan

.github/workflows/website.yml

@@ -62,8 +62,23 @@ jobs:
     runs-on: ubuntu-latest
     environment:
       name: github-pages
-      url: ${{ steps.deployment.outputs.page_url }}
+      url: ${{ steps.deployment.outputs.page_url || steps.deployment-retry.outputs.page_url }}
     steps:
       - name: Deploy
         id: deployment
+        continue-on-error: true
+        uses: actions/deploy-pages@v5
+
+      # The Pages backend intermittently rejects deployments with a generic
+      # "Deployment failed, try again later". When that hits the post-release
+      # workflow_run deploy, the appcast silently goes stale and Sparkle stops
+      # offering updates until an unrelated push redeploys the site. Retry once
+      # after a pause before failing the run.
+      - name: Wait before retry
+        if: steps.deployment.outcome == 'failure'
+        run: sleep 60
+
+      - name: Retry deploy
+        id: deployment-retry
+        if: steps.deployment.outcome == 'failure'
         uses: actions/deploy-pages@v5