fix(stacks): configure ollama for stability and persistent model loading

- OLLAMA_NUM_PARALLEL=1: Process requests sequentially to avoid batch
  overflow panics when multiple services call embeddings concurrently
- OLLAMA_KEEP_ALIVE=-1: Keep models loaded in VRAM indefinitely
- OLLAMA_MAX_LOADED_MODELS=2: Allow both mistral-nemo and nomic-embed-text
  to stay loaded simultaneously

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

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

stacks/ollama.yml

@@ -22,6 +22,12 @@ services:
       - TZ=Europe/Amsterdam
       - NVIDIA_VISIBLE_DEVICES=all
       - NVIDIA_DRIVER_CAPABILITIES=all
+      # Process requests sequentially to avoid batch overflow panics
+      - OLLAMA_NUM_PARALLEL=1
+      # Keep models loaded in VRAM (don't unload after idle)
+      - OLLAMA_KEEP_ALIVE=-1
+      # Load multiple models concurrently (mistral-nemo + nomic-embed-text)
+      - OLLAMA_MAX_LOADED_MODELS=2
     deploy:
       resources:
         limits: