fix(rtx6000): enable scale-to-zero with 20min cooldown + wake-up route

- Set replicas.min: 0 for scale-to-zero when idle
- Add config.cooldownPeriod: 1200 to KEDA HTTP add-on HelmRelease
  (20 min buffer before scaling down from 1 replica)
- Add external HTTPRoute at brain.siriusdevops.com for manual wake-up
  via curl from Hermes/Telegram

infrastructure/controllers/base/keda/release-http-add-on.yaml

@@ -19,3 +19,6 @@ spec:
     crds: CreateReplace
   dependsOn:
     - name: keda
+  values:
+    config:
+      cooldownPeriod: 1200

infrastructure/gpus/base/keda-gpu-scaling/keda-vllm.yaml

@@ -13,6 +13,6 @@ spec:
     service: rtx6000-brain-service
     port: 8000
   replicas:
-    min: 1
+    min: 0
     max: 1
  

infrastructure/gpus/base/keda-gpu-scaling/kustomization.yaml

@@ -3,4 +3,5 @@ apiVersion: kustomize.config.k8s.io/v1beta1
 kind: Kustomization
 resources:
   - keda-vllm.yaml
+  - vllm-route.yaml
 

infrastructure/gpus/base/keda-gpu-scaling/vllm-route.yaml

@@ -0,0 +1,19 @@
+
+apiVersion: gateway.networking.k8s.io/v1
+kind: HTTPRoute
+metadata:
+  name: rtx6000-brain-route
+  namespace: customer1
+spec:
+  parentRefs:
+  - name: external-http-gateway
+  hostnames:
+  - brain.siriusdevops.com
+  rules:
+  - matches:
+    - path:
+        type: PathPrefix
+        value: /
+    backendRefs:
+    - name: rtx6000-brain-service
+      port: 8000