test: vector_search E2E with real embeddings + LangChain similarity_search

[polaz]

DB gate: G073 ✅ (VectorService → HNSW wiring complete)

Task

• E2E tests for vector_search() with real float embeddings via Docker
• Verify similarity_search() in LangChain adapter returns ranked results

[coderabbitai]

⚠️ Possible Duplicate Issue(s)

• fix: vector_search() always returns empty — E2E tests + LangChain similarity_search() #20

🔗 Related PRs

#27 - feat: similarity_search() for LangChain + upsert_relations() idempotency test [merged]

---

📝 Issue Planner

_{Check the box below or use the @coderabbitai plan command to generate an implementation plan and prompts that you can use with your favorite coding assistant.}

• Create Plan

---

🧪 Issue enrichment is currently in open beta.

You can configure auto-planning by selecting labels in the issue_enrichment configuration.

To disable automatic issue enrichment, add the following to your .coderabbit.yaml:

issue_enrichment:
  auto_enrich:
    enabled: false

💬 Have feedback or questions? Drop into our discord!

[polaz]

Реализовано. E2E-тесты с реальными float-эмбеддингами существуют:

• LlamaIndex: test_vector_query_returns_results — сеет :Chunk с уникальным 16-мерным вектором, проверяет что seeded node присутствует в top-k
• LangChain: test_similarity_search_returns_results — проверяет ranked порядок (distance[i] <= distance[i+1])
  Закрываю как выполненное.