[Feature]: turboquant: KV cache

[qdrddr]

I'd like to benefit from KV Cache quantization on macOS
https://github.com/OnlyTerp/turboquant
https://github.com/mitkox/vllm-turboquant
https://github.com/scrya-com/rotorquant
https://github.com/VectorDB-NTU/RaBitQ-Library

[joepio]

https://github.com/arozanov/turboquant-mlx

[the-wondersmith]

@qdrddr I took a pass at getting turboquant added (specifically using the turboquant-mlx package @joepio suggested), and put up a PR for it. I'm only passingly familiar with mlx-engine's internals though, so it may or may not pass muster.

The ball is at least rolling now though, so it's something 😅