feat(knowledgebase): replace litellm-pgvector connector with memory-retrieval

Per docs/research/langchain-pgvector-vs-litellm-pgvector.md (issue #25):
the vendored litellm-pgvector connector (793 lines, Prisma migrations, a
fragile git-context build) is replaced by a ~90-line FastAPI service
(services/memory-retrieval/) wrapping langchain_postgres.PGVector directly
against pgvector-db. Same gateway boundary — it still calls litellm for
embeddings, nothing talks to Postgres or the model directly except this
service.

- New services/memory-retrieval/ (main.py, Dockerfile, requirements.txt):
  POST /ingest, POST /query, GET /health.
- docker-compose.yml: litellm-pgvector service replaced by memory-retrieval;
  pgvector-db and embedding-server untouched.
- litellm-config.yaml: vector_store_registry block removed (no
  langchain_postgres provider exists to register against; callers query
  memory-retrieval directly instead of an in-band file_search tool call —
  that mechanism was never confirmed working per issue #24 anyway).
- scripts/ingest-memory.sh rewritten for the new /ingest endpoint (same
  per-line chunking, no dedup).
- .env vars renamed: LITELLM_PGVECTOR_API_KEY/LITELLM_PGVECTOR_EMBEDDING_KEY
  -> MEMORY_RETRIEVAL_API_KEY/MEMORY_RETRIEVAL_EMBEDDING_KEY.
- vendor/litellm-pgvector/ removed entirely.
- docs/memory-knowledgebase.md updated for the new setup/query flow.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

vendor/litellm-pgvector/config.py

@@ -1,60 +0,0 @@
-from typing import Dict, Optional
-from pydantic import BaseModel
-from pydantic_settings import BaseSettings
-
-
-class DatabaseFieldConfig(BaseModel):
-    """Configuration for database field mappings"""
-    id_field: str = "id"
-    content_field: str = "content"
-    metadata_field: str = "metadata"
-    embedding_field: str = "embedding"
-    vector_store_id_field: str = "vector_store_id"
-    created_at_field: str = "created_at"
-
-
-class EmbeddingConfig(BaseModel):
-    """Configuration for embedding generation via LiteLLM proxy"""
-    model: str = "text-embedding-ada-002"
-    base_url: str = "http://localhost:4000"  # LiteLLM proxy URL
-    api_key: str = "sk-1234"  # LiteLLM proxy API key
-    dimensions: int = 1536
-
-
-class Settings(BaseSettings):
-    """Application settings"""
-    # Database configuration
-    database_url: str = "postgresql://username:password@localhost:5432/vectordb?schema=public"
-    
-    # API configuration
-    server_api_key: str = "your-api-key-here"
-    port: int = 8000
-    host: str = "0.0.0.0"
-    
-    # Database field mappings
-    db_fields: DatabaseFieldConfig = DatabaseFieldConfig()
-    
-    # Embedding configuration
-    embedding: EmbeddingConfig = EmbeddingConfig()
-    
-    class Config:
-        env_file = ".env"
-        env_nested_delimiter = "__"
-        case_sensitive = False
-        
-        # Allow environment variables like:
-        # DB_FIELDS__ID_FIELD=custom_id
-        # EMBEDDING__MODEL=text-embedding-3-small
-        # EMBEDDING__API_BASE=https://api.openai.com/v1
-        
-    @property
-    def table_names(self) -> Dict[str, str]:
-        """Get table names"""
-        return {
-            "vector_stores": "vector_stores",
-            "embeddings": "embeddings"
-        }
-
-
-# Global settings instance
-settings = Settings()