[Doc]: Failed to download lora adapter using the path from documentation

[Jeffwan]

📚 The doc issue

https://docs.vllm.ai/en/latest/models/lora.html describe the steps to load a lora model.

python -m vllm.entrypoints.openai.api_server \
    --model meta-llama/Llama-2-7b-hf \
    --enable-lora \
    --lora-modules sql-lora=~/.cache/huggingface/hub/models--yard1--llama-2-7b-sql-lora-test/

There're two issues

1. The model path is incorrect. We should append snapshots/0dfa347e8877a4d4ed19ee56c140fa518470028c

2. ~ is not expanded automatically and it fails to load the model, at this moment, relative path is not supported.

Screenshots

1. Path documented

2. Update the path with appended snapshot commit id

3. Update to absolute path

Suggest a potential alternative/fix

1. append commit it snapshots/0dfa347e8877a4d4ed19ee56c140fa518470028c
2. change ~ to $HOME

[Jeffwan]

I will submit a PR for short-term fix and separate PR to support ~ and dynamic loading from model registery.