Any plans in adding support for DeepSpeedHybridEngine

Nobody has claimed this yet.

### Description & Motivation

I'm using lightning for my project using PPO algorithm.
During an optimisation step, both inference and training are needed, so it's slow with LLMs.
Recently, DeepSpeed introduced a HybridEngine for this scenario, with the benefits quoted below:
> The Hybrid-Engine is capable of seamlessly transitioning between inference and training modes within RLHF, allowing it to leverage various optimizations from DeepSpeed-Inference such as tensor-parallelism and high-performance transformer kernels for generation...

### Pitch

I wonder if you have any plan to add support for this new hybrid engine?

### Alternatives

A pointer to tutorials on how to add custom engines/strategies are also much appreciated.
I can create a PR afterwards if my project allows.

### Additional context

Link to [DeepSpeedHybridEngine](https://github.com/microsoft/DeepSpeed/blob/d1c3c0df5318cbcb2b221d38d4127dd566f4984d/deepspeed/runtime/hybrid_engine.py#L32)

cc @borda @awaelchli

Start by reading DeepSpeed's runtime/hybrid_engine.py, linked in the issue, and review the project's guidance for adding custom engines or strategies. Define the integration scope and acceptance criteria for switching between inference and training during PPO optimization; the issue does not name Lightning files or tests.

Written by the indexing model from the issue text.