Disable {save,check}_on_train_epoch_end with check_val_every_n_epoch>1

[carmocca]

What does this PR do?

Fixes #9151, #9163

Does your PR introduce any breaking changes? If yes, please list them.

None

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?
• [n/a] 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 internal minor changes/refactorings)

PR review

• 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

[ananthsub]

@carmocca do you think it'd be better to have the user explicitly opt into this without providing a default?

[codecov]

Codecov Report

Merging #9156 (23dbc5a) into master (b497fb8) will increase coverage by 0%.
The diff coverage is 100%.

@@          Coverage Diff           @@
##           master   #9156   +/-   ##
======================================
  Coverage      92%     92%           
======================================
  Files         176     178    +2     
  Lines       14804   14850   +46     
======================================
+ Hits        13651   13701   +50     
+ Misses       1153    1149    -4     

[carmocca]

No, the problem is that users don't notice something is wrong until the epoch is over and a callback tries to get a non-logged metric key.

We could print a warning before the epoch starts asking to set this flag when trainer.val_check_interval != 1.0 or trainer.check_val_every_n_epoch != 1, but then they'd ask: "Why don't you set it automatically?" which is what we do here.

[tchaton]

LGTM !