Fix resetting epoch loop restarting flag in LearningRateFinder

[clumsy]

What does this PR do?

lr_finder does not reset fit_loop.epoch_loop.restarting to False which causes the validation check to trigger validation on the first advance() call because of this check.

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.
Before you start reviewing, make sure you have read the review guidelines. In short, see the following bullet-list:

Reviewer checklist

• Is this pull request ready for review? (if not, please submit in draft mode)
• Check that all items from Before submitting are resolved
• Make sure the title is self-explanatory and the description concisely explains the PR
• Add labels and milestones (and optionally projects) to the PR so it can be classified

---

📚 Documentation preview 📚: https://pytorch-lightning--19819.org.readthedocs.build/en/19819/

[clumsy]

Could you please check this fix @awaelchli , @carmocca ?

[awaelchli]

Thank you for the fix @clumsy, very nice!
It looks like the batch size finder would need the same fix as well. Are you interested to investigate it as well?

[clumsy]

Interesting, batch_size_finder actually worked for me but I can check and make the change, sure.

There was a more critical bug for LR finder, I'm afraid, @awaelchli: #19814

[clumsy]

This flag reset is seem to be in place for batch_size_finder here, @awaelchli