rewrite waterdynamics to use AnalysisBase

[orbeckst]

Is your feature request related to a problem?

Code in analysis.waterdynamics does not behave like most of the analysis tools that ship with MDAnalysis.

There are various inconsistencies, notably

• Make waterdynamics stop behaviour the same as AnalysisBase #22
• waterdynamics does NOT respect PBC #19
• use AtomGroup in place of selection strings mdanalysis#2369 / Refactor waterdynamics to use UpdatingAtomGroup and add benchmark. #20

and the current approach also does not allow us to easily parallelize (#4080 ).

Describe the solution you'd like

Refactor code to use AnalysisBase, see how to write your own analysis code

Describe alternatives you've considered

Do nothing and maintain the code as is (but do not consider any enhancements or improvements).

Additional context

We are also considering decided moving wateranalysis to a separate MDAKit. Once the code is on a separate development track, it may be easier for users/developers to contribute.

MDAnalysis/mdanalysis#4310

[orbeckst]

cc: @alejob @ianmkenney @IAlibay @fiona-naughton @lilyminium

[orbeckst]

See MDAnalysis/mdanalysis#4310 .

This issue may eventually be moved to a separate MDAKit repository.

[orbeckst]

@ALescoulie are you still interested in working on waterdynamics? If so, it's now in a stable MDAKit state with a release and ready to support new innovation or just good ol' refactoring.

[ALescoulie]

I'm interested, just been occupied working on other projects, hoping to get started on it soon after I finish my personal portfolio website and some MD-SAPT stuff.