crosslink RAG Blueprint Elasticsearch, Pinecone, and Teradata VDB guides

[kheiss-uwzoo]

PR title

docs: crosslink RAG Blueprint Elasticsearch, Pinecone, and Teradata VDB guides

Summary

Extends docs/docs/extraction/vdbs.md so readers can find Elasticsearch (NVIDIA RAG Blueprint), Pinecone (nvidia-pinecone-rag), and Teradata (TeradataVDB) alongside the in-library LanceDB / OpenSearch content.

Changes

• Add subsection RAG Blueprint and partner vector stores with a short explanation of how these paths differ from the default NeMo Retriever Library + LanceDB flow.
• Add a crosslink table:
  • Configure Elasticsearch as Your Vector Database for NVIDIA RAG Blueprint
  • Customize your vector database (Pinecone + NVIDIA RAG)
  • TeradataVDB (NVIDIA NIM Ingest integration)
• Clarify that testing and release cadence for those integrations follow the owning project, consistent with existing first-party LanceDB guidance.
• Update the on this page TOC with the new anchor.

Out of scope

• No link for IBM EnterpriseDB (no confirmed public URL).
• No Milvus source link (stable public implementation URL not verified).

[greptile-apps]

Greptile Summary

This PR extends docs/docs/extraction/vdbs.md with a new ### RAG Blueprint and partner vector stores subsection and updates the on-page TOC to match.

• Adds a crosslink table pointing readers to external configuration guides for Elasticsearch (NVIDIA RAG Blueprint), Pinecone (pinecone-io/nvidia-pinecone-rag), and Teradata (TeradataVDB), together with a note that release cadence for those integrations follows the owning project.
• Updates the "On this page" TOC with the new anchor #rag-blueprint-and-partner-vector-stores.

Confidence Score: 5/5

Documentation-only change; no code paths are touched, and all new content is additive crosslinks with no risk to existing functionality.

The change is purely additive Markdown in a single docs file — no logic, API surface, or configuration is modified. The new section correctly scopes ownership and testing responsibility to external projects, and the existing LanceDB guidance is untouched.

No files require special attention.

Important Files Changed

Filename	Overview
docs/docs/extraction/vdbs.md	Adds "RAG Blueprint and partner vector stores" subsection with crosslinks for Elasticsearch, Pinecone, and Teradata; minor anchor-format inconsistency and a pre-existing structural concern about the !!! important admonition scope.

Flowchart

%%{init: {'theme': 'neutral'}}%%
flowchart TD
    A[NeMo Retriever Library] --> B[LanceDB\ndefault / first-party]
    A --> C[OpenSearch\nreference VDB adapter]
    A --> D[NVIDIA RAG Blueprint]
    A --> E[Partner packages]
    D --> F[Elasticsearch\ndocs.nvidia.com/rag]
    E --> G[Pinecone\npinecone-io/nvidia-pinecone-rag]
    E --> H[Teradata TeradataVDB\ndocs.teradata.com]
    style B fill:#c8e6c9
    style F fill:#fff9c4
    style G fill:#fff9c4
    style H fill:#fff9c4

Loading

Prompt To Fix All With AI

Fix the following 1 code review issue. Work through them one at a time, proposing concise fixes.

---

### Issue 1 of 1
docs/docs/extraction/vdbs.md:164
The new heading uses `{#rag-blueprint-and-partner-vector-stores}` (no spaces), while every other custom anchor in this file uses the spaced form `{ #anchor-name }`. MkDocs Material's `attr_list` extension accepts both, but inconsistent style makes future automated anchor extraction or linting fragile.

```suggestion
### RAG Blueprint and partner vector stores { #rag-blueprint-and-partner-vector-stores }
```

_{Reviews (3): Last reviewed commit: "Merge branch 'main' into kheiss/elastics..." | Re-trigger Greptile}