common : add -hfd option for the draft model

[ggerganov]

Same as -hf arg, but for the draft model.

Example

llama-server \
    -hf  ggml-org/DeepSeek-R1-Distill-Qwen-32B-Q8_0-GGUF \
    -hfd ggml-org/DeepSeek-R1-Distill-Qwen-1.5B-Q4_0-GGUF \
    --ctx-size 32768 -fa -t 1 --draft-max 16 --draft-min 2