Provide a resource usage overview for Operators

[viccuad]

Several people have asked for specifics WRT resource usage for operators.
Current docs provide operators on how to deploy, but they would need to deploy and/or read helm chart values to see what is the expected resource usage, and how it grows.

We should document this instead. By documenting this, we also showcase that we have resource usage in mind.

Acceptance criteria

Add a Resource usage page in the Operator manual docs.

Currently, the helm charts set resource limits and requests for each container image, which should be more than enough. Highlight those and point to them in the docs. E.g: https://github.com/kubewarden/helm-charts/blob/main/charts/kubewarden-controller/values.yaml#L118-L132

Explain briefly the Kubewarden stack, starting from the resource angle. What resources do we need for the default deployment. How does it grow vertically (instantiating more policies into a PolicyServer) and horizontally (instantiating more PolicyServers). Mention replicas for all Deployments. Mention needed stateful components (none so far, no DBs, no PVs).

Mention briefly latency WRT to validating and mutating webhooks, and k8s apiserver queries.

Add how to achieve HA:

• Mention replicas for all deployments
• Mention that for PolicyServers, the resulting deployments are managed by the controller. But the controller patches only the parts it needs on the deployment. This means that it is possible to manuall set affinity tolerations, requests, limits for the policy-server deployments.