Logger support in Lite

[lightningforever]

What does this PR do?

This PR adds the logger base class and TensorBoardLogger implementation to Fabric.

logger = TensorBoardLogger()
fabric = LightningLite(loggers=logger)

fabric.log("accuracy", acc)
fabric.log_dict({"accuracy": acc, "loss": loss})

The base class interface is mostly the same as in PyTorch Lightning, but with these key differences:

• Removed model checkpoint callback dependency and after_save_checkpoint hook. Lite won't have the same checkpointing mechanism through callbacks, and thus we will have to rework the way loggers access checkpoints.
• Removed DummyLogger (only needed for Trainer)
• Moved properties to the top (code organization)
• Property: save_dir -> root_dir. We are standardizing the way the log dir is defined, to be compatible with as many logger frameworks as possible. More info in Introduce Logger.experiment_dir #14188.

The TensorBoard implementation changes very little. Trainer-only features and dependencies were removed:

• Removed hparams saving
• Removed OmegaConf dependency
• save_dir -> root_dir
• Removed log_graph argument in constructor (logging graph is still supported through logger.log_graph())

Future Work:

• Support for self.log in LightningModule will be added in WIP Add support for self.log from LightningModule in Lite #16135

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

No

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?

I made sure I had fun coding 🙃

[tchaton]

LGMT !