Update OceanBaseStore to be compatible with oceanbase-4.3.3.0

[GITHUBear]

Description

OceanBase officially released the vector store function in version 4.3.3. The SQL statements to use vector index have been changed. For compatibility, the implementation of OceanBaseStore should be updated:

• Using pyobvector as middleware
• Support filters when perform ANN search

How Has This Been Tested?

• Test the knowledge base function using OceanBase + DB-GPT.
• make fmt
• make mypy
• make test

Snapshots:

Checklist:

• My code follows the style guidelines of this project
• I have already rebased the commits and make the commit message conform to the project standard.
• I have performed a self-review of my own code
• I have commented my code, particularly in hard-to-understand areas
• I have made corresponding changes to the documentation
• Any dependent changes have been merged and published in downstream modules

[csunny]

Hi @GITHUBear, Can the default value of ob_vector_memory_limit_percentage be set to 30?

[SQL: CREATE VECTOR INDEX vidx ON obvector_test_v2 (embedding) WITH (distance=l2,m=16,ef_construction=256,lib=vsag,type=hnsw)]

[csunny]

Cool, works and easy to use than previous versions.

[GITHUBear]

Can the default value of ob_vector_memory_limit_percentage be set to 30?

@csunny Yes, we recommend setting the value of ob_vector_memory_limit_percentage to 30. This can be done via pyobvector:

from pyobvector import ObVecClient

tmp_client = ObVecClient()
tmp_client.perform_raw_text_sql(
    "ALTER SYSTEM ob_vector_memory_limit_percentage = 30"
)

Since a tenant only needs to set ob_vector_memory_limit_percentage once, we do not involve it into pyobvector.

[csunny]

What I mean is that OB sets the default value of ob_vector_memory_limit_percentage in the kernel parameters to 30. 😊
@GITHUBear

[csunny]

LGTM, 🎉🎉

[Aries-ckt]

LGTM