split restore_training_state into logical parts [2 / 2]

[awaelchli]

What does this PR do?

Part of #7652 and
follow up to #7901 will be updated after #7901 merged!

Splits restore_training_state into the following pieces:

restore_callbacks()
restore_progress()
restore_optimizers()
restore_lr_schedulers()

Before submitting

• Was this discussed/approved 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 update the CHANGELOG? (not for typos, docs, test updates, or internal minor changes/refactorings)

PR review

Anyone in the community is free to review the PR once the tests have passed.
Before you start reviewing make sure you have read Review guidelines. In short, see the following bullet-list:

• 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

Did you have fun?

Make sure you had fun coding 🙃

[codecov]

Codecov Report

Merging #7900 (5d65f4b) into master (d209b68) will decrease coverage by 0%.
The diff coverage is 88%.

@@          Coverage Diff           @@
##           master   #7900   +/-   ##
======================================
- Coverage      92%     92%   -0%     
======================================
  Files         200     200           
  Lines       12837   12816   -21     
======================================
- Hits        11857   11832   -25     
- Misses        980     984    +4     

[tchaton]

Neat !

[ananthsub]

when can this happen?

[awaelchli]

when Trainer(resume_from_checkpoint=None)
(the default case)
then trainer calls restore() and we do a no-op

[ananthsub]

shouldn't we skip calling restore at all in that case?

[awaelchli]

no because it's actually an attempt to restore.
this is the sequence:

1. attempt to restore from hpc_checkpoint file
2. attempt to restore from resume_from_checkpoint path if provided
3. do nothing

and could rename restore to attempt_restore if that's better?