Is your improvement request related to a feature? Please describe (👍 if you like this request)

The conversion/admission webhook and recovery backend services are deployed as Deployment. Despite the anti-affinity for these services, there's still a chance that these deployments' replicas are on the same worker node. In the event that a worker node fails due to a network partition or a power outage, the services will be irresponsive until their pods are recreated on different worker nodes. The single node failure will result in the whole Longhorn cluster getting stuck. To achieve high availability, it is better to deploy these three services by DaemonSet rather than Deployment, as they are critical components for resource operation (conversion/admission webhooks) and RWX volume operation (recovery backend).

Describe the solution you'd like

A clear and concise description of what you want to happen.

Describe alternatives you've considered

A clear and concise description of any alternative solutions or features you've considered.

Additional context

Add any other context or screenshots about the feature request here.