Invalid cross-device link

Recently, this workflow is giving errors when used on AWS self-hosted GPU runners:

Example: https://github.com/openvm-org/stark-backend/actions/runs/17200888597/job/48791119926?pr=102

Error snippet:
```
error: could not rename component file from '/home/runner/.rustup/toolchains/stable-x86_64-unknown-linux-gnu/share/doc/clippy' to '/home/runner/.rustup/tmp/qzwl4905618fqaq1_dir/bk': Invalid cross-device link (os error 18)
```

Internet/LLM searches say GPU instances are virtualized differently, but the same workflows used to run fine until 2 days ago, so I wonder if something has changed in the workflow to cause this?

I now add
```
- run: |
rustup component remove clippy || true
rm -rf ~/.rustup/toolchains/stable-x86_64-unknown-linux-gnu || true
```
prior to
```
- uses: dtolnay/rust-toolchain@stable
```
in all workflows to fix this.