[DOCUMENTATION] ROCm install using Docker

[PonyLucky]

Is your feature request related to a problem? Please describe.
No.

Describe the solution you'd like
A line on README.md to explain how to install LocalAI using Docker on AMD GPUs (maybe a second one for our intel friends...).

In the Or run with docker:.

Describe alternatives you've considered
After a search on Dockerhub I see (here) 3 main categories:

1. nvidia-cuda - NVIDIA GPUs.
2. intel - Intel ^^.
3. hipblas - When I installed ROCm locally once I saw HIP somewhere so I guess AMD.

If my suposition is right, can you add:

# Intel GPU:
docker run -ti --name local-ai -p 8080:8080 --gpus all localai/localai:latest-gpu-intel-f32

# AMD GPU
docker run -ti --name local-ai -p 8080:8080 --gpus all localai/localai:latest-gpu-hipblas

in the README please.

Additional context
I checked if AMD GPUs are supported first, it seems to work (#1592).

[jtwolfe]

I'll try to make some adjustments to the docs for this, there are a few inadequacies in the current docs regarding other functionality that i want to address too so I'll just add this to my list.