Refactor sub modules

[RNKuhns]

Reference Issues/PRs

Partial implementation of skbase design proposal.

What does this implement/fix? Explain your changes.

Refactors skbase functionality into skbase.base, skbase.lookup, skbase.validate. Presence of skbase.testing remains unchanged.

[RNKuhns]

@fkiraly note that I've removed the comment from lgtm-com because we are using CodeQL now (that is name of lgtm functionality that was ported into Github). We need to disable lgtm-com app integration.

[codecov-commenter]

Codecov Report

Merging #78 (494cc93) into main (d61ae40) will increase coverage by 1.27%.
The diff coverage is 83.89%.

@@            Coverage Diff             @@
##             main      #78      +/-   ##
==========================================
+ Coverage   57.85%   59.12%   +1.27%     
==========================================
  Files          16       22       +6     
  Lines        1350     1397      +47     
==========================================
+ Hits          781      826      +45     
- Misses        569      571       +2     

Impacted Files	Coverage Δ
skbase/validate/_types.py	45.83% <45.83%> (ø)
skbase/_exceptions.py	77.77% <66.66%> (-22.23%)	⬇️
skbase/base/_meta.py	16.34% <81.81%> (ø)
skbase/utils/_nested_iter.py	92.85% <92.85%> (ø)
skbase/__init__.py	100.00% <100.00%> (ø)
skbase/base/__init__.py	100.00% <100.00%> (ø)
skbase/base/_base.py	81.38% <100.00%> (ø)
skbase/lookup/__init__.py	100.00% <100.00%> (ø)
skbase/lookup/_lookup.py	43.84% <100.00%> (ø)
skbase/mock_package/mock_package.py	100.00% <100.00%> (ø)
... and 6 more

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

[RNKuhns]

@fkiraly I believe this has refactored everything into the different modules as noted in the design proposal in #76. Should be ready for your review.

[fkiraly]

Makes much sense.

I suppose this is the sktime equivalent of sorting your garden gnome collection into different cabinets and then arranging them in regular geometric patterns. We can all relate.

Some non-blocking comments:

• should _remove_single also be in _nested_seq? It unnests.
• _nested_seq, perhaps better _nested_iter or _nested_list (knowing it also applies to tuples)?

[RNKuhns]

Makes much sense.

I suppose this is the sktime equivalent of sorting your garden gnome collection into different cabinets and then arranging them in regular geometric patterns. We can all relate.

I like that analogy! And yep, that's the plan, get everything in an order so it is easier to add the rest of the functionality from here.

[RNKuhns]

• should _remove_single also be in _nested_seq? It unnests.
  That does make sense.

• _nested_seq, perhaps better _nested_iter or _nested_list (knowing it also applies to tuples)?
  I think that _nested_iter makes sense b/c tuples and lists are both iterators. I actually plan to cleanup a bit of this functionality in a later PR, and make some of that a bit more generic to other iterators, so your suggested naming convention works with that.

I'll include both updates in a commit, then merge after the tests pass.