chore: Enable keyword search for Milvus inline

[varshaprasad96]

What does this PR do?

With milvus-io/milvus-lite#294 - Milvus Lite supports keyword search using BM25. While introducing keyword search we had explicitly disabled it for inline milvus. This PR removes the need for the check, and enables inline::milvus for tests.

Test Plan

Run llama stack with inline::milvus enabled:

pytest tests/integration/vector_io/test_openai_vector_stores.py::test_openai_vector_store_search_modes --stack-config=http://localhost:8321 --embedding-model=all-MiniLM-L6-v2 -v

INFO     2025-08-07 17:06:20,932 tests.integration.conftest:64 tests: Setting DISABLE_CODE_SANDBOX=1 for macOS                                        
=========================================================================================== test session starts ============================================================================================
platform darwin -- Python 3.12.11, pytest-7.4.4, pluggy-1.5.0 -- /Users/vnarsing/miniconda3/envs/stack-client/bin/python
cachedir: .pytest_cache
metadata: {'Python': '3.12.11', 'Platform': 'macOS-14.7.6-arm64-arm-64bit', 'Packages': {'pytest': '7.4.4', 'pluggy': '1.5.0'}, 'Plugins': {'asyncio': '0.23.8', 'cov': '6.0.0', 'timeout': '2.2.0', 'socket': '0.7.0', 'html': '3.1.1', 'langsmith': '0.3.39', 'anyio': '4.8.0', 'metadata': '3.0.0'}}
rootdir: /Users/vnarsing/go/src/github/meta-llama/llama-stack
configfile: pyproject.toml
plugins: asyncio-0.23.8, cov-6.0.0, timeout-2.2.0, socket-0.7.0, html-3.1.1, langsmith-0.3.39, anyio-4.8.0, metadata-3.0.0
asyncio: mode=Mode.AUTO
collected 3 items                                                                                                                                                                                          

tests/integration/vector_io/test_openai_vector_stores.py::test_openai_vector_store_search_modes[None-None-all-MiniLM-L6-v2-None-384-vector] PASSED                                                   [ 33%]
tests/integration/vector_io/test_openai_vector_stores.py::test_openai_vector_store_search_modes[None-None-all-MiniLM-L6-v2-None-384-keyword] PASSED                                                  [ 66%]
tests/integration/vector_io/test_openai_vector_stores.py::test_openai_vector_store_search_modes[None-None-all-MiniLM-L6-v2-None-384-hybrid] PASSED                                                   [100%]

============================================================================================ 3 passed in 4.75s =============================================================================================

[varshaprasad96]

@franciscojavierarceo a quick one. With this we will have feature parity between inline and remote milvus for search modes.

[franciscojavierarceo]

Nice! I approved the workflow, let's wait for the tests to pass.

[franciscojavierarceo]

Shouldn't you update pyproject.toml for the latest release? Or do we already have the latest version that supports bm25?

[franciscojavierarceo]

Looks like we must have the latest version. Thanks for the quick turnaround on this!!!

[leseb]

We should really have a constraint on the minimal required version for milvus-lite. +1 on #3073 (comment)

[varshaprasad96]

@leseb @franciscojavierarceo The reason Milvus-lite version was updated, was because we are locking pymilvus>=2.5 (https://github.com/meta-llama/llama-stack/blob/1677d6bffdf0a002abe3b827c460df4930ee83c8/uv.lock#L1746) and Milvus-lite is a transitive dependency. I've also updated to pin the Milvus-lite version to be sure that we are importing the one which has this change implemented.

[ChristianZaccaria]

/lgtm nice!!

[varshaprasad96]

@leseb could we get this merged this, please? We would need it for upcoming release.

[bbrowning]

The changes look reasonable, self-contained, tested, and feedback looks to be addressed. Thanks!

Dismissing since Seb's request was addressed. 👍