fix: vLLM deployment — Recreate strategy, correct service port, single model name

- Strategy: Recreate (GPU can only be used by one pod at a time)
- nuextract-external targetPort: 8080 (nginx proxy port)
- served-model-name: numind/NuExtract3
- DB ai_agents updated to model=numind/NuExtract3 provider=vllm

infra/kube-vllm/deployment.yaml

@@ -7,6 +7,8 @@ metadata:
     app: vllm
 spec:
   replicas: 1
+  strategy:
+    type: Recreate
   selector:
     matchLabels:
       app: vllm

infra/kube-vllm/service.yaml

@@ -23,7 +23,7 @@ spec:
     - nodePort: 31508
       name: vllm-api
       port: 2701
-      targetPort: 8000
+      targetPort: 8080
   selector:
     app: vllm-metrics
   type: LoadBalancer