fix(linstor): increase satellite startup probe failure threshold (#2425)

## What this PR does

Increases the LINSTOR satellite startup probe `failureThreshold` from
the Kubernetes default of 3 (30 seconds) to 30 (300 seconds / 5
minutes).

The default startup probe is too aggressive for nodes with slow storage
initialization — satellites get killed and restarted before they have a
chance to come up. This adds an explicit `startupProbe` override in the
`LinstorSatelliteConfiguration` pod template, keeping the same
`tcpSocket` check on the `linstor` port but giving satellites
significantly more time to start.

### Release note

```release-note
fix(linstor): increase satellite startup probe failure threshold from 3 to 30 (5 minutes to start)
```

<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit

* **Chores**
* Added a startup health check for the LINSTOR satellite container to
improve pod reliability and ensure the service fully initializes before
accepting traffic.
* This change only introduces container startup probing behavior and
does not alter other runtime behavior or public interfaces.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->

packages/system/linstor/templates/satellites-cozy.yaml

@@ -14,6 +14,11 @@ spec:
       containers:
       - name: linstor-satellite
         image: {{ .Values.piraeusServer.image.repository }}:{{ .Values.piraeusServer.image.tag }}
+        startupProbe:
+          tcpSocket:
+            port: linstor
+          periodSeconds: 10
+          failureThreshold: 30
         securityContext:
           # real-world installations need some debugging from time to time
           readOnlyRootFilesystem: false