[doc] Add RAG Integration example

[reidliu41]

RAG (Retrieval-Augmented Generation) enhances LLMs by retrieving relevant context from external sources,
improving factual accuracy and grounding. It's widely adopted in modern LLM applications.
This PR introduces basic RAG example using:

• vLLM + LangChain + Milvus
• vLLM + LlamaIndex + Milvus

[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.

🚀

[DarkLight1337]

cc @hmellor there seems to be some indentation errors even though this PR doesn't change them

[hmellor]

Those indentation errors (which should be solved) are harmless for the build.

The main issue is that:

:::{argparse}
:module: examples.online_serving.retrieval_augmented_generation_with_langchain
:func: get_parser
:prog: retrieval_augmented_generation_with_langchain.py
:::

requires the imports from that module to either be mocked (docs/source/conf.py) or installed (requirements/docs.txt).

[reidliu41]

@hmellor yeah, seems that, thanks
@DarkLight1337 seems cannot import, maybe rollback the previous command output? seems not good to change some settings/configs for the examples.

[DarkLight1337]

We can add langchain and llamaindex to the dependencies to mock inside conf.py

[reidliu41]

[2025-05-06T13:03:22Z] Warning, treated as error:
--
  | [2025-05-06T13:03:22Z] /vllm-workspace/test_docs/docs/source/deployment/frameworks/retrieval_augmented_generation.md:42:Failed to import "get_parser" from "examples.online_serving.retrieval_augmented_generation_with_langchain".
  | [2025-05-06T13:03:22Z] No module named 'examples.online_serving'
  | [2025-05-06T13:03:45Z] make: *** [Makefile:20: html] Error 2
  | [2025-05-06T13:03:46Z] 🚨 Error: The command exited with status 2

still failed...

[reidliu41]

@DarkLight1337 maybe just simply remove it or just roll back??

[DarkLight1337]

I prefer removing the help text if you can't get it to work, so we don't have to worry about the help text getting out of sync with the actual code

[reidliu41]

ok, thanks

[DarkLight1337]

Thanks for your effort and patience!