fix(k8s): use Recreate strategy for neuron-mcp deployments

Single-node k3s cluster cannot schedule a second pod during rolling update
(512Mi request × 2 exceeds available memory). Recreate terminates the old
pod before starting the new one, trading brief downtime for schedulability.

servers/legion/k8s/neuron-technologies/prod/mcp-deployment.yaml

@@ -8,6 +8,8 @@ metadata:
     env: prod
 spec:
   replicas: 1
+  strategy:
+    type: Recreate
   selector:
     matchLabels:
       app: neuron-mcp

servers/legion/k8s/neuron-technologies/stage/mcp-deployment.yaml

@@ -8,6 +8,8 @@ metadata:
     env: stage
 spec:
   replicas: 1
+  strategy:
+    type: Recreate
   selector:
     matchLabels:
       app: neuron-mcp