perf: optimize Docker build time from >10min to ~2min

- Remove sentence-transformers dependency (saves ~3GB PyTorch/CUDA)
- Make CrossEncoder import optional with graceful fallback
- Optimize Dockerfile for layer caching (incremental builds ~3s)
- Change PostgREST port from 3000 to 3001 (avoid Next.js conflict)

🤖 Generated with [Claude Code](https://claude.com/claude-code)

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

servers/mcp-crawl4ai-rag/pyproject.toml

@@ -10,6 +10,5 @@ dependencies = [
     "supabase==2.15.1",
     "openai==1.71.0",
     "dotenv==0.9.9",
-    "sentence-transformers>=4.1.0",
     "neo4j>=5.28.1",
 ]