Don't combine separate hub/proxy/etc pods cpu/memory metrics in related panels

[consideRatio]

Hub cpu/memory panels should only show one hub pod I think, not sum or average multiple pods values. Currently they are summed over, which is why we see the hub pod reported to have 2GB of memory here.

Selecting All hubs (namespaces really) in a cluster

Selecting one hub (namespace really) in a cluster