Switch embedding model to bge-m3 (1024-dim)

The Qdrant collections use 1024-dim vectors (bge-m3) but the
embedding-service was configured with all-MiniLM-L6-v2 (384-dim).
Also increase memory limit to 8G for the larger model.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

docker-compose.yml

@@ -414,7 +414,7 @@ services:
       - embedding_models:/root/.cache/huggingface
     environment:
       EMBEDDING_BACKEND: ${EMBEDDING_BACKEND:-local}
-      LOCAL_EMBEDDING_MODEL: ${LOCAL_EMBEDDING_MODEL:-sentence-transformers/all-MiniLM-L6-v2}
+      LOCAL_EMBEDDING_MODEL: ${LOCAL_EMBEDDING_MODEL:-BAAI/bge-m3}
       LOCAL_RERANKER_MODEL: ${LOCAL_RERANKER_MODEL:-cross-encoder/ms-marco-MiniLM-L-6-v2}
       PDF_EXTRACTION_BACKEND: ${PDF_EXTRACTION_BACKEND:-pymupdf}
       OPENAI_API_KEY: ${OPENAI_API_KEY:-}
@@ -423,7 +423,7 @@ services:
     deploy:
       resources:
         limits:
-          memory: 4G
+          memory: 8G
     healthcheck:
       test: ["CMD", "python", "-c", "import httpx; r=httpx.get('http://127.0.0.1:8087/health'); r.raise_for_status()"]
       interval: 30s