Expanding support for the solution of bilevel optimization problems in JuMP.

This GSOC will add support for the solution of a much wider class of bilevel optimization problems to JuMP. Currently, support for bilevel optimization is through BilevelJuMP.jl, which is primarily capable of solving problems for which there is a compact, strong dual for the lower-level problem. The aim of this project is to allow the integer variables and possibly even stochasticity to be included at both upper and lower levels. For this purpose, we propose two approaches. 1- Integrate the well-developed open-source solver into BilevelJuMP.jl. 2- As a stretch goal, extend Dualization.jl to include concepts from the duality of integer programming.