Fix dtype for MPS in reinforcement learning example

[swyo]

What does this PR do?

Fixes #19981

Before submitting

• Was this discussed/agreed via a GitHub issue? (not for typos and docs)
• Did you read the contributor guideline, Pull Request section?
• Did you make sure your PR does only one thing, instead of bundling different changes together?
• Did you make sure to update the documentation with your changes? (if necessary)
• Did you write any new necessary tests? (not for typos and docs)
• Did you verify new and existing tests pass locally with your changes?
• Did you list all the breaking changes introduced by this pull request?
• Did you update the CHANGELOG? (not for typos, docs, test updates, or minor internal changes/refactors)

PR review

Anyone in the community is welcome to review the PR.
The only part I am modifying is the example code, so there are no other dependencies affected by this change.

---

📚 Documentation preview 📚: https://pytorch-lightning--19982.org.readthedocs.build/en/19982/

[awaelchli]

Thanks 🚀