Remove training_trick_connector.py

[daniellepintz]

What does this PR do?

Fixes #10108

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

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 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

Anyone in the community is welcome to review the PR.
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 🙃

[rohitgr7]

I know there isn't much code in some of these connectors but moving all of them to trainer isn't a good idea IMO. Makes the trainer more conjested. I mean right now these are clustered in their own classes which makes it easy to add new features or do refactors. But after these let's say something new comes in future then again we will think of moving these to separate classes.

[daniellepintz]

@rohitgr7 I hear your point. Right now we are only moving the very small connectors to the trainer or elsewhere, not the larger ones like data_connector, accelerator_connector, etc.

Regarding making the trainer more congested please read #8946 (comment)

I think @ananthsub makes a great point there which is, the trainer is currently a god class with a lot of complication. Keeping that complication in different connectors doesn't actually fix the problem, but just moves it out of sight. To start improving things we can first consolidate the code in the trainer and then slowly simplify it over time. We are only doing this for very small connectors right now, for large connectors we will come up with a well designed plan for them.

[codecov]

Codecov Report

Merging #10112 (aaf795e) into master (3a4e997) will decrease coverage by 4%.
The diff coverage is 100%.

@@           Coverage Diff            @@
##           master   #10112    +/-   ##
========================================
- Coverage      92%      89%    -4%     
========================================
  Files         180      179     -1     
  Lines       16146    16136    -10     
========================================
- Hits        14933    14307   -626     
- Misses       1213     1829   +616