Limit ollama to one loaded model and one parallel request

Prevents OOM when switching between models — only one 14B model
fits in 16GB VRAM at a time with KV cache for context.

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

argocd/manifests/ollama/deployment.yaml

@@ -28,6 +28,10 @@ spec:
               value: /models
             - name: OLLAMA_HOST
               value: "0.0.0.0:11434"
+            - name: OLLAMA_MAX_LOADED_MODELS
+              value: "1"
+            - name: OLLAMA_NUM_PARALLEL
+              value: "1"
           volumeMounts:
             - name: models
               mountPath: /models