add input_embeddings to LLM interface

[davidkoski]

See https://github.com/ml-explore/mlx-swift-examples/pull/238/files

The gemma3_text model has an input_embeddings parameter:

• https://github.com/ml-explore/mlx-lm/blob/main/mlx_lm/models/gemma3_text.py#L224

Per the python docs on generate_step:

        input_embeddings (mx.array, optional): Input embeddings to use in place of
           prompt tokens. Default: ``None``.

mlx-swift-examples doesn't support this -- should it?