ci(docker): retry image build to absorb transient Docker Hub registry errors (#41069)

Co-authored-by: Claude Code <noreply@anthropic.com>

.github/workflows/docker.yml

@@ -101,13 +101,27 @@ jobs:
             PUSH_OR_LOAD="--load"
           fi
 
-          supersetbot docker \
-            $PUSH_OR_LOAD \
-            --preset "$BUILD_PRESET" \
-            --context "$EVENT" \
-            --context-ref "$RELEASE" $FORCE_LATEST \
-            --extra-flags "--build-arg INCLUDE_CHROMIUM=false --tag $IMAGE_TAG" \
-            $PLATFORM_ARG
+          # Retry to absorb transient Docker Hub registry errors (base-image
+          # pull timeouts, 504/401 on push, ECONNRESET) that otherwise fail
+          # the whole job. buildx reuses the buildkit layer cache from the
+          # failed attempt, so a retry mostly re-does just the failed push.
+          for attempt in 1 2 3; do
+            if supersetbot docker \
+              $PUSH_OR_LOAD \
+              --preset "$BUILD_PRESET" \
+              --context "$EVENT" \
+              --context-ref "$RELEASE" $FORCE_LATEST \
+              --extra-flags "--build-arg INCLUDE_CHROMIUM=false --tag $IMAGE_TAG" \
+              $PLATFORM_ARG; then
+              break
+            fi
+            if [ "$attempt" -eq 3 ]; then
+              echo "::error::supersetbot docker build failed after 3 attempts"
+              exit 1
+            fi
+            echo "::warning::Build attempt ${attempt} failed; retrying in 30s..."
+            sleep 30
+          done
 
       # in the context of push (using multi-platform build), we need to pull the image locally
       - name: Docker pull