fix: update port manager to use multiprocessing lock for process safety

[deependujha]

What does this PR do?

PortManager should be process safe, as we run distributed tests by launching new process in a batch of 5.

Addresses comment: #21309 (comment)

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--21311.org.readthedocs.build/en/21311/

[deependujha]

PL has a regression test test_import_lightning_multiprocessing_start_method_not_set, which fails if we either do multiprocess.Lock() or use manager.Lock() in global scope, as it sets multiprocessing context.

But, instantiating them in a function or class (Lazy instantiation) will make it process specific and won't be shared across different processes to make it process-safe.

I can't think of other way except file-based port tracking to make it process safe.

[Borda]

Does it need to be for all and so could we have it as a wrapper?

[littlebullGit]

#21313
@Borda @deependujha I think I address the multiprocessing issue using File locks. Please take a look of the above PR