update doc for llama3

[zhyncs]

Motivation

as titled

• Add llama3 chat template #1461
• warning transformers version #1453

ref #1459

Modification

as titled

Checklist

1. Pre-commit or other linting tools are used to fix the potential lint issues.
2. The modification is covered by complete unit tests. If not, please add more unit tests to ensure the correctness.
3. If the modification has a dependency on downstream projects of a newer version, this PR should be tested with all supported versions of downstream projects.
4. The documentation has been modified accordingly, like docstring or example tutorials.

[AllentDan]

LGTM

[zhyncs]

The compatibility verification is all OK with main branch.

# TurboMind
python3 -m lmdeploy serve api_server /workdir/Meta-Llama-3-8B

# PyTorch
python3 -m lmdeploy serve api_server /workdir/Meta-Llama-3-8B --backend pytorch

# KV Cache Int8
python3 -m lmdeploy serve api_server /workdir/Meta-Llama-3-8B --quant-policy 8

# KV Cache Int4
python3 -m lmdeploy serve api_server /workdir/Meta-Llama-3-8B --quant-policy 4

# AWQ
python3 -m lmdeploy lite auto_awq /workdir/Meta-Llama-3-8B --calib-dataset 'ptb' --calib-samples 128 --calib-seqlen 2048 --w-bits 4 --w-group-size 128 --work-dir /workdir/Meta-Llama-3-8B-AWQ
python3 -m lmdeploy serve api_server /workdir/Meta-Llama-3-8B-AWQ

# press
python3 benchmark/profile_restful_api.py --server_addr 127.0.0.1:23333 --tokenizer_path /workdir/Meta-Llama-3-8B --dataset /workdir/ShareGPT_V3_unfiltered_cleaned_split.json --concurrency 128 --num_prompts 1000