deprecate _update_called

[matsumotosan]

What does this PR do?

Removes deprecated Metric._update_called in favor of Metric.update_called.

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--2141.org.readthedocs.build/en/2141/

[SkafteNicki]

Here are the lines that should be updated on lightning side:
https://github.com/Lightning-AI/lightning/blob/09a0fb26d2bb819c6cbabf87c695b657ac5d305f/src/lightning/pytorch/trainer/connectors/logger_connector/result.py#L268-L273

[awaelchli]

Thank you for taking care of this. With Lightning-AI/pytorch-lightning#18714 Lightning 2.1 will call the public method. Sorry that it was calling the protected API 😅

What is torchmetrics' policy around supporting different lightning versions? Could one use a newer torchmetrics version with an older lightning version? If yes, probably the change here would have to wait otherwise it would break. Anyway, great work on sorting it out for the future! Thanks

[SkafteNicki]

What is torchmetrics' policy around supporting different lightning versions? Could one use a newer torchmetrics version with an older lightning version? If yes, probably the change here would have to wait otherwise it would break. Anyway, great work on sorting it out for the future! Thanks

@awaelchli I do not think we have an official policy regarding lightning, only torch where we aim for new releases of torchmetrics to be compatible with up to 2 year old releases of torch.
@Borda opinion on what we should do here?

[Borda]

I do not think we have an official policy regarding lightning

As we are the same group, we try to support as large as we can without too much overhead, but said so, we do not need to be more generous than lightning itself meaning we are actively maintaining 1.9.x and 2.0+ even here we have set range:

torchmetrics/requirements/integrate.txt

Line 1 in d8962b4

pytorch-lightning >=1.6.0, <2.1.0

also addressed in #2129

So, to be clear, this would be a breaking change such that users with TM 1.3+ can't use lighting bellow 2.1, correct?
cc: @lantiga

[codecov]

Codecov Report

Merging #2141 (8b26630) into master (d8a7a55) will increase coverage by 0%.
The diff coverage is 100%.

Additional details and impacted files

@@          Coverage Diff           @@
##           master   #2141   +/-   ##
======================================
  Coverage      87%     87%           
======================================
  Files         293     293           
  Lines       16396   16397    +1     
======================================
+ Hits        14252   14253    +1     
  Misses       2144    2144