Boundary Conditions as Collage-Decapodes

[lukem12345]

Closes #144

With @p-stokes , this branch will encode boundary conditions and initial conditions as morphisms between Decapode diagrams.

One Decapode represents the physical equations, and the other represents values to set along boundaries.

TODO:

• Add docs page
• Pass simple masking test with a discrete boundary Decapode
• Initial conditions

Pass masking test where a Var in the discrete Decapode is computed from other vars in that Decapode.

• This feature is "extra", and so lies out of the scope of this PR.

[codecov]

Codecov Report

Attention: 1 lines in your changes are missing coverage. Please review.

Comparison is base (3395d01) 75.17% compared to head (6de2efd) 76.13%.

❗ Current head 6de2efd differs from pull request most recent head c3c9e3a. Consider uploading reports for the commit c3c9e3a to get more accurate results

Files	Patch %	Lines
src/simulation.jl	93.33%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #151      +/-   ##
==========================================
+ Coverage   75.17%   76.13%   +0.96%     
==========================================
  Files          11       10       -1     
  Lines        1172     1207      +35     
==========================================
+ Hits          881      919      +38     
+ Misses        291      288       -3     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[lukem12345]

The collate function as of the most recent commit is correct for boundary masks applied to state variables or tangent variables.

[lukem12345]

I have marked this PR as "ready for review" since the source code and tests are complete (up to passing the review stage). So, the code and tests themselves are ready to review as I concurrently write the docs and example scripts to accompany this new feature set.
@jpfairbanks @p-stokes

[jpfairbanks]

Sadly this didn't get merged before the component-arraypocalypse. so it needs to be brought up to date (@lukem12345). There are also merge conflicts now.

[lukem12345]

Disadvantage to not using findnode as a thin wrapper around getfield, but such is git.

[lukem12345]

What are your thoughts on the docs page. Good enough for merge? The accompanying demo is the so-called brusselator_bounded.jl example. @jpfairbanks

[lukem12345]

DiagrammaticEquations.jl has the collage source and test code now, so this PR is closed by that repo!