[Feat][Frontend] Added support for HermesToolParser for models without special tokens

[minpeter]

Upstream the implementation implemented in the https://github.com/minpeter/hermes-llama-parse repo,

Modify the hermes parser to be compatible with models that do not have separate "<tool_call>", "</tool_call>" tags (e.g. llama),

*I know there is a llama-specific parser, but it is useful in fine-tuning llama-based models.

[github-actions]

👋 Hi! Thank you for contributing to the vLLM project.

💬 Join our developer Slack at https://slack.vllm.ai to discuss your PR in #pr-reviews, coordinate on features in #feat- channels, or join special interest groups in #sig- channels.

Just a reminder: PRs would not trigger full CI run by default. Instead, it would only run fastcheck CI which starts running only a small and essential subset of CI tests to quickly catch errors. You can run other CI tests on top of those by going to your fastcheck build on Buildkite UI (linked in the PR checks section) and unblock them. If you do not have permission to unblock, ping simon-mo or khluu to add you in our Buildkite org.

Once the PR is approved and ready to go, your PR reviewer(s) can run CI to test the changes comprehensively before merging.

To run CI, PR reviewers can either: Add ready label to the PR or enable auto-merge.

🚀

[aarnphm]

Is there a test fine-tuned model that we can use to test this?

maybe a very small finetune llama3.2-1b would work here.

[minpeter]

Is there a test fine-tuned model that we can use to test this?

maybe a very small finetune llama3.2-1b would work here.

https://huggingface.co/minpeter/LoRA-Llama-3b-xlam-vllm-test-ci

[aarnphm]

Can you add a testcase using that model?

[minpeter]

Can you add a testcase using that model?

Let me try, is there any other test code I can refer to?

[aarnphm]

iirc there are a few tool parser tests under tests/entrypoints/openai/tool_parsers