Should this package extend NetworkLayouts.jl?

[asinghvi17]

That would make it a drop-in replacement for graph layout algorithms in e.g GraphMakie.jl. The only change that would be needed are:

• The solvers inherit from NetworkLayout.AbstractLayout
• The solver structs are themselves callable and return vectors of GeometryBasics.Point2

For this, we'd have to add three new fields to the Zarate struct that correspond to the kwargs in the current solve_positions function. But this wouldn't technically be breaking from a user perspective, since that function's kwargs can still override the solver specification if desired.

What do you think @oxinabox? I'm happy to put in a PR if you think this makes sense.

[oxinabox]

I think that makes sense.
NetworkLayouts is fairly lightweight.

Maybe we want to make a struct (ZarateLayout) that wraps the current struct.
So existing uses don't have to change at all.

I would review such a PR