TODOs for achieving stability for Trainer Strategy API

[kaushikb11]

Proposed refactoring or deprecation

Motivation

We have added the strategy argument for Trainer in this PR.

The next immediate refactors required are:

• Tests (1/n) tests: Use strategy flag instead of accelerator for training strategies #9931
• Code/warnings updates Update accelerator connector messages after the addition of strategy #9937
• Docs: general usage Update strategy flag in docs #10000
• Docs: Trainer flag documentation (accelerator, strategy) Update Trainer flag docs for strategy #10042
• tutorials

Post the above refactors, we will be going through the below internal updates:

• Remove DistributedType.DDP_CPU. This should not be a valid strategy value.
• Remove DistributedType.TPU_SPAWN
• Rename DistributedType to StrategyType Deprecate DistributedType in favor of StrategyType #10505
• Rename DeviceType to AcceleratorType Deprecate DeviceType in favor of _AcceleratorType #10503
• Remove user-facing distributed_backend references in code and error messages. Remove deprecated distributed_backend from Trainer #10017
• Rename the inernal self.distributed_backend references.
• Rename AcceleratorConnector._device_type to _accelerator_type
• Rename AcceleratorConnector._distrib_type to _strategy_type
• Define all non-supported accelerator-strategy combinations. Is this already done?
• Rename TrainingTypePlugin and its references to StrategyPlugin Introduce Strategy in favour of TrainingTypePlugin #10548
• Remove "test_accelerator_x" tests in tests/accelerators/test_accelerator_connector.py that test for the deprecated usage of accelerator="strategy_name"

---

If you enjoy Lightning, check out our other projects! ⚡

• Metrics: Machine learning metrics for distributed, scalable PyTorch applications.

• Flash: The fastest way to get a Lightning baseline! A collection of tasks for fast prototyping, baselining, finetuning and solving problems with deep learning

• Bolts: Pretrained SOTA Deep Learning models, callbacks and more for research and production with PyTorch Lightning and PyTorch

• Lightning Transformers: Flexible interface for high performance research using SOTA Transformers leveraging Pytorch Lightning, Transformers, and Hydra.

[daniellepintz]

Do we want to also rename DeviceType here to AcceleratorType?
https://github.com/PyTorchLightning/pytorch-lightning/blob/7b4df7bf919acfd7f7b39d780faeadb54aec9ade/pytorch_lightning/utilities/enums.py#L114

[carmocca]

Added to the list.