Add disk cleanup for Docker builds (#1442)

<!-- .github/pull_request_template.md -->

## 📌 Description

<!-- What does this PR do? Briefly describe the changes and why they’re
needed. -->

## 🔍 Related Issues

<!-- Link any related issues here -->

## 🚀 Pull Request Checklist

Thank you for contributing to FlashInfer! Before we review your pull
request, please make sure the following items are complete.

### ✅ Pre-commit Checks

- [ ] I have installed `pre-commit` by running `pip install pre-commit`
(or used your preferred method).
- [ ] I have installed the hooks with `pre-commit install`.
- [ ] I have run the hooks manually with `pre-commit run --all-files`
and fixed any reported issues.

> If you are unsure about how to set up `pre-commit`, see [the
pre-commit documentation](https://pre-commit.com/).

## 🧪 Tests

- [ ] Tests have been added or updated as needed.
- [ ] All tests are passing (`unittest`, etc.).

## Reviewer Notes

<!-- Optional: anything you'd like reviewers to focus on, concerns, etc.
-->

Author: yongwww

.github/workflows/release-ci-docker.yml

@@ -7,6 +7,17 @@ jobs:
   build:
     runs-on: ubuntu-latest
     steps:
+      - name: Free Disk Space
+        uses: jlumbroso/free-disk-space@main
+        with:
+          tool-cache: false
+          android: true
+          dotnet: true
+          haskell: true
+          large-packages: true
+          docker-images: true
+          swap-storage: true
+
       - uses: actions/checkout@v4
 
       - name: Set up QEMU
@@ -21,15 +32,47 @@ jobs:
           username: flashinfer
           password: ${{ secrets.DOCKERHUB_TOKEN }}
 
-      - name: Build and push multi-arch image
+      - name: Build and push AMD64 image
         uses: docker/build-push-action@v5
         with:
           context: docker
           file: docker/Dockerfile.ci_gpu
-          platforms: linux/amd64,linux/arm64
+          platforms: linux/amd64
           push: true
           tags: |
-            flashinfer/flashinfer-ci:latest
-            flashinfer/flashinfer-ci:${{ github.sha }}
-          cache-from: type=registry,ref=flashinfer/flashinfer-ci:buildcache
-          cache-to: type=registry,ref=flashinfer/flashinfer-ci:buildcache,mode=max
+            flashinfer/flashinfer-ci:amd64
+          cache-from: type=registry,ref=flashinfer/flashinfer-ci:buildcache-amd64
+          cache-to: type=registry,ref=flashinfer/flashinfer-ci:buildcache-amd64,mode=max
+          provenance: false
+          sbom: false
+
+      - name: Clean up after AMD64 build
+        run: |
+          df -h /
+          docker buildx prune -af
+          docker system prune -af --volumes
+          sudo rm -rf /tmp/* /var/tmp/* ~/.docker/buildx
+          df -h /
+
+      - name: Re-setup Docker Buildx for ARM64
+        uses: docker/setup-buildx-action@v3
+
+      - name: Build and push ARM64 image
+        uses: docker/build-push-action@v5
+        with:
+          context: docker
+          file: docker/Dockerfile.ci_gpu
+          platforms: linux/arm64
+          push: true
+          tags: |
+            flashinfer/flashinfer-ci:arm64
+          cache-from: type=registry,ref=flashinfer/flashinfer-ci:buildcache-arm64
+          cache-to: type=registry,ref=flashinfer/flashinfer-ci:buildcache-arm64,mode=max
+          provenance: false
+          sbom: false
+
+      - name: Create and push multi-arch manifest
+        run: |
+          docker buildx imagetools create -t flashinfer/flashinfer-ci:latest \
+            flashinfer/flashinfer-ci:amd64 \
+            flashinfer/flashinfer-ci:arm64