What would you like to be added:
gardenctl should be able to fetch configuration, like the target patterns (#44) or e.g. access restriction configuration from the garden cluster.

Gardenctl could then fetch ConfigMap(s) from a predefined namespace, like gardenctl-system with configuration data for gardenctl.

Gardenctl will also read the cluster-identity configmap in kube-system namespace from the garden cluster and will use this for the garden configuration, e.g. it will store the value under .gardens[].clusterIdentity. The .gardens[].name field could be dropped in favour of clusterIdentity. We could also think about duplicating the cluster-identity from the kube-system namespace to the gardenctl-system namespace so that also non-operator users are able to get this information.

A dedicated command could be introduced for this purpose, like gardenctl config sync [--garden landscape-dev | --dry-run]. If no garden cluster identity or alias is provided, all configured gardens are synced.

Why is this needed:
This is needed so that the user only needs to provide minimal configuration data and the tool fetches the rest of the configuration automatically.