feat(cli): aggregate entities that differs only by a certain property or parameterization

[danielelotito]

I faced this problem:

I have an experiment parameterized by n_seeds, I might have in different spaces measured entities that differ only by n_seeds. n_seeds represents the number of times a certain stochastic algorithm has been repeated.
The other parameterization parameters and constitutive properties of the entity determine the configuration space of my stochastic algorithm
This is exactly what happens for the cplex_mip custom experiment

I wanted to know a simple way to query the samplestore to know which configurations have been repeated the highest (and 1rst to highest, 2nd to highest,... Kth to highest) number of times.
I can return the top K

Example
As an example, I consider a samplestore with 3 spaces some space1 configurations are also in space2 and some other space1 configurations are also in space3, and n_seeds(space1)>n_seeds(space2)>n_seeds(space3).

The output.csv for K=2 could be something such as

     [all the columns I get from show entities], space_id, sum_n_seeds_value
     [some_measured_entities_from_space1], space1, total_n_seeds_highest
...
     [some_measured_entities_from_space2], space2, total_n_seeds_highest
...
     [some_measured_entities_from_space1], space1, total_n_seeds_1rst_to_highest
...
     [some_measured_entities_from_pace3], space3, total_n_seeds_1rst_to_highest
...
...