Fix support for half precision in Perplexity metric

[SkafteNicki]

What does this PR do?

Fixes #2224

Before submitting

• Was this discussed/agreed via a Github issue? (no need for typos and docs improvements)
• Did you read the contributor guideline, Pull Request section?
• Did you make sure to update the docs?
• Did you write any new necessary tests?

PR review

Anyone in the community is free to review the PR once the tests have passed.
If we didn't discuss your PR in Github issues there's a high chance it will not be merged.

Did you have fun?

Make sure you had fun coding 🙃

---

📚 Documentation preview 📚: https://torchmetrics--2235.org.readthedocs.build/en/2235/

[codecov]

Codecov Report

Merging #2235 (e67d254) into master (9c964ec) will decrease coverage by 40%.
Report is 1 commits behind head on master.
The diff coverage is 50%.

Additional details and impacted files

@@           Coverage Diff            @@
##           master   #2235     +/-   ##
========================================
- Coverage      87%     47%    -40%     
========================================
  Files         293     293             
  Lines       16425   16423      -2     
========================================
- Hits        14278    7724   -6554     
- Misses       2147    8699   +6552     

[Borda]

this looks strange:

>       assert time_cg < time_no_cg, "using compute groups were not faster"
E       AssertionError: using compute groups were not faster
E       assert 24.99868607521057 < 0.33540940284729004