Explore (and or automate) global optimization

Choose a system where we believe that global optimization will be important. This means a system where we suspect that there are lots of possible ground states and convergence issues are likely to arise. Two possible systems which could be considered:

• strained graphene, (already have the kwant code from the ~qah-strained-graphene project), know to expect convergence issues.
• bilayer graphene as system to explore (system with lots of symmetries possible, meaning that there are likely a whole bunch of ground states possible.

After choosing a system, or at least with a specific system in mind, explore different global optimization algorithms. Explore whether minimizing wrt to energy is sufficient or whether a different cost function is better. If it turns out that the procedure is somewhat standard, we can also automate it and interface it into the code.