Interface to MathOptInterface

[matbesancon]

This will allow us to access the predefined sets, projections from MathOptSetDistances.jl.

From a user point of view, this means one can potentially use the algorithms from the package with the https://jump.dev interface

[pokutta]

would be great to interface with jump!

[pokutta]

especially the projections we might need for bcg or if we want to add fcfw

[matbesancon]

especially the projections we might need for bcg or if we want to add fcfw

so you mean for this using JuMP as a sub-solver for LMO?

[pokutta]

so there is two ways actually:

1. use jump to formulate and solve the lmo. this will add some overhead but better than interfacing scip, glop, etc all separately.
2. BCG and FCFW require the solution of additional subproblems that are not LPs but require some projections onto simple sets, in fact probability simplices -> i saw that you already implemented many projections, so that we can use them for that

[pokutta]

but also the other around is interesting: using this package as a solver within jump

so i guess it goes both ways

[pokutta]

wait until nonlinear interface in jump is stable.