docs: align dense index lifecycle documentation

[LHT129]

Summary

Align dense-index lifecycle and input-type documentation with the current implementation.

Changes

• Document BruteForce UpdateVector support and its FP32 input contract.
• Document FP32 input boundaries for Pyramid, IVF, and BruteForce.
• Correct lifecycle coverage for Pyramid and IVF MARK_REMOVE support.
• Document BruteForce deletion restrictions with attribute filtering and multi-vector mode.

Validation

• Ran git diff --check.
• Checked the corrected capability statements and the Chinese/English document pairs.
• No C++ code or tests changed.

Related

• Related to [Docs] Align dense index lifecycle documentation with implementation #2761

[vsag-bot]

/label status/waiting-for-review
/waiting-on reviewer
/request-review @jiaweizone
/request-review @wxyucs
/request-review @inabao

[mergify]

Merge Protections

🟢 All 2 merge protections satisfied — ready to merge.

Show 2 satisfied protections

🟢 Require kind label

• label~=^kind/

🟢 Require version label

• label~=^version/

[Copilot]

Pull request overview

Updates the VSAG documentation to match current dense-index lifecycle behavior and FP32 input contracts, addressing the capability mismatches called out in #2761.

Changes:

• Clarifies that Pyramid/IVF/BruteForce public dense-vector paths accept FP32 input via Dataset::Float32Vectors, and that internal quantizers do not expand accepted input types.
• Corrects lifecycle/capability documentation: Pyramid and IVF support RemoveMode::MARK_REMOVE only; BruteForce supports UpdateVector and has documented deletion restrictions.
• Aligns Chinese and English dense-index docs for the same capability statements and caveats.

Reviewed changes

Copilot reviewed 8 out of 8 changed files in this pull request and generated no comments.

Show a summary per file

File	Description
docs/docs/zh/src/indexes/pyramid.md	Adds FP32-only input contract clarification for Pyramid.
docs/docs/zh/src/indexes/ivf.md	Adds FP32-only input contract clarification + notes IVF rejects dtype: "int8".
docs/docs/zh/src/indexes/brute_force.md	Documents FP32-only input, UpdateVector support, and deletion restrictions (attribute filter / multi-vector).
docs/docs/zh/src/advanced/index_lifecycle.md	Updates lifecycle/capability matrix and deletion/update semantics across indexes.
docs/docs/en/src/indexes/pyramid.md	Adds FP32-only input contract clarification for Pyramid.
docs/docs/en/src/indexes/ivf.md	Adds FP32-only input contract clarification + notes IVF rejects dtype: "int8".
docs/docs/en/src/indexes/brute_force.md	Documents FP32-only input, UpdateVector support, and deletion restrictions (attribute filter / multi-vector).
docs/docs/en/src/advanced/index_lifecycle.md	Updates lifecycle/capability matrix and deletion/update semantics across indexes.

---

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.

[Copilot]

Pull request overview

Copilot reviewed 8 out of 8 changed files in this pull request and generated 4 comments.

[Copilot]

Pull request overview

Copilot reviewed 8 out of 8 changed files in this pull request and generated 3 comments.

[Copilot]

Pull request overview

Copilot reviewed 8 out of 8 changed files in this pull request and generated 2 comments.

[LHT129]

[suggestion] The per-index brute_force.md (both EN and ZH) deletion sections document the use_attribute_filter restriction but omit the multi-vector mode restriction that is correctly documented in index_lifecycle.md. The two pages should be consistent.

In docs/docs/en/src/indexes/brute_force.md line 131 and docs/docs/zh/src/indexes/brute_force.md line 124, only the use_attribute_filter restriction is noted. The index_lifecycle.md (both EN and ZH) correctly adds:

When operating in multi-vector mode, BruteForce supports RemoveMode::MARK_REMOVE only; FORCE_REMOVE is not available.

Consider adding a similar note to the per-index deletion sections so readers consulting either page get the complete picture.