[FEATURE] Add total memory information for each node

[amitgalitz]

Is your feature request related to a problem? Currently our profile API gives us the total_size_in_bytes for each detector, however not the total memory used by AD in each node and the entire cluster. We also currently don't have much metrics on our cache like hit:miss ratio other than active entities.

What solution would you like? We can add either to the profile API or stats API the total memory used by AD in each node and the entire cluster by adding up the size in bytes for each detector