[v1.38] Document Boost API

[g-despot]

Summary

Documents the v1.38 Preview Boost API — a post-retrieval rescorer that soft-promotes or demotes search results without removing non-matching documents.

Companion to server PR weaviate/weaviate#11103 and Python client PR weaviate/weaviate-python-client#2030.

What's in here

• New how-to: docs/weaviate/search/boost.md — condition types (filter / property / time decay / numeric decay), curves, blending, depth, negative weights, hybrid behavior. Python-only tabs.
• Snippet: _includes/code/howto/search.boost.py — seven worked examples, verified end-to-end against weaviate:1.38.0-rc.0 + text2vec-transformers via tests/docker-compose-anon.yml.
• Cross-links: new ## Soft-rank with Boost sections on similarity.md, hybrid.md, bm25.md, rerank.md, all using a shared <BoostPreview/> admonition.
• Env var: QUERY_BOOST_DEFAULT_DEPTH added to the env-vars reference.
• Sidebar: weaviate/search/boost inserted after rerank.
• Compose bump: every tests/docker-compose*.yml upgraded from weaviate:1.37.2 → weaviate:1.38.0-rc.0.

Test plan

• yarn build clean.
• Snippet runs end-to-end against weaviate:1.38.0-rc.0 with text2vec-transformers.

[orca-security-eu]

Orca Security Scan Summary

Status	Check	Issues by priority
Passed	Infrastructure as Code	0   0   0   0	View in Orca
Passed	SAST	0   0   0   0	View in Orca
Passed	Secrets	0   0   0   0	View in Orca
Passed	Vulnerabilities	0   0   0   0	View in Orca

[orca-security-eu]

Orca Security Scan Summary

Status	Check	Issues by priority
Passed	Secrets	0   0   0   0	View in Orca

[Copilot]

Pull request overview

Documents the v1.38 Preview Boost API (post-retrieval soft reranking) and integrates it into existing search documentation so users can discover and apply Boost across vector, BM25, hybrid, and rerank-related workflows.

Changes:

• Adds a new how-to page for Boost (concepts, condition types, curves, blending/weights, and depth/pagination).
• Adds a shared “Preview” admonition include and cross-links (“Soft-rank with Boost”) from existing search pages.
• Adds a Python end-to-end snippet and documents the QUERY_BOOST_DEFAULT_DEPTH environment variable; updates the sidebar to include the new page.

Reviewed changes

Copilot reviewed 9 out of 9 changed files in this pull request and generated 2 comments.

Show a summary per file

File	Description
sidebars.js	Adds the Boost doc page to the “How-to: Query & Search” sidebar.
docs/weaviate/search/similarity.md	Adds Boost preview import and a “Soft-rank with Boost” section linking to the new page.
docs/weaviate/search/rerank.md	Adds Boost preview import and guidance on when to use Boost vs reranking modules.
docs/weaviate/search/hybrid.md	Adds Boost preview import and clarifies Boost runs on the fused hybrid results.
docs/weaviate/search/boost.md	New Boost how-to page detailing behavior, condition types, curves, blending, and depth.
docs/weaviate/search/bm25.md	Adds Boost preview import and a “Soft-rank with Boost” section linking to the new page.
docs/deploy/configuration/env-vars/index.md	Documents QUERY_BOOST_DEFAULT_DEPTH.
_includes/feature-notes/boost.mdx	New reusable preview admonition for Boost.
_includes/code/howto/search.boost.py	New Python snippet with multiple Boost examples used by the new docs page.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.