Rest API (beta): Add model size field to models/ routes

[havenwood]

Feature request: Add a size (or byte_size) field representing model size for GET /api/v0/models and GET /api/v0/models/{model} response bodies.

Current:

{
  "id": "qwen2-vl-7b-instruct",
  "object": "model",
  "type": "vlm",
  "publisher": "mlx-community",
  "arch": "qwen2_vl",
  "compatibility_type": "mlx",
  "quantization": "4bit",
  "state": "not-loaded",
  "max_context_length": 32768
}

Feature request:

{
  "id": "qwen2-vl-7b-instruct",
  "object": "model",
  "type": "vlm",
  "publisher": "mlx-community",
  "arch": "qwen2_vl",
  "compatibility_type": "mlx",
  "quantization": "4bit",
  "state": "not-loaded",
  "max_context_length": 32768
  "size": 11596411699
}

Model size can be a fairly important factor, for a variety of reasons. It's also a common field to use for sorting models. I think the model byte size would be a nice addition to the models/ and modes/{model} GET routes.