[Refactor Suggestions II] Functional, with some inheritance

[CodyCBakerPhD]

First of several draft proposals opened for comparison; each offers advantages and disadvantages in terms of readability and comprehension (dependent on knowledge of package structure)

This one centralizes duplicate code into common functions, including some base benchmarks where appropriate

Reduces redirection compared to #19, but still requires some understanding of the base benchmark definitions to understand what tests are being run in the cases where a benchmark inherits

Also beginning to adopt sklearn import structure since centralized imports are now being used

[oruebel]

Just a couple of things I noticed while working on the network benchmarks:

1. When using setup_cache, ASV appear to run the cache setup only once for the base class, but does not run it again for child classes that inherit from that class.
2. We wouldn't want to run time_slice in the base class, I assume adding the underscore prefix in the py filename is to prevent ASV from detecting the file

[CodyCBakerPhD]

1. When using setup_cache, ASV appear to run the cache setup only once for the base class, but does not run it again for child classes that inherit from that class.

Great to know; I'd consider this a point in favor of #21 since it removes the need for having to know that implicit behavior of ASV

2. We wouldn't want to run time_slice in the base class, I assume adding the underscore prefix in the py filename is to prevent ASV from detecting the file

Related to #19 (comment); more important that it be outside of the nwb_benchmarks/benchmarks testing directory

The prepended underscore is just sklearn import style that effectively says 'the thing I want to publicly expose is the class AcquisitionTimeSeriesSliceBenchmark, not the arbitrarily named submodule it resides in' (which is purely for ease of navigability within the project)