Create low barrier to entry UIs for Geometry Central

[keenancrane]

There are a lot of great algorithms in GC, but it's sometimes annoying to have to build a whole C++ project, or hack on some "spaghetti" project, when you just want to do a few operations to a mesh, or a batch of meshes. (As you might want to do during, say, a paper deadline! :-P)

Doesn't seem like it would be too hard to hack together various ways of chaining together outputs and inputs:

• a node-based interface for hooking up GC algorithms (e.g., via React Flow and Electron)
• a collection of command line utilities for invoking GC algorithms, that support UNIX "piping" between utilities
• and/or a command line REPL for processing meshes via GC (probably best implemented by just exposing more of Geometry Central through potpourri3d)

None of these utilities belong in the geometry-central repo itself, since they're effectively just apps built on top of GC. However, the question is: is there any software engineering we can do within GC to grease the wheels? I.e., to make it easier to chain inputs to outputs? Or are we basically already there?

[This issue is mostly just a note to myself.]