Setting the stage for full Bayesian optimization

[dflemin3]

This PR adds a few quality of life improvements and is prepping approxposterior for full-on Bayesian optimization.

• I generalized the minimizeObjective function to be a more general minimizer and now use it in the findMAP function
• Added the Jones et al. (1998) utility (aka acquisition) function for Bayesian optimization
• Generalized the findNextPoint function to find an arbitrary number of points
• approxposterior can now handle 1D functions
• Removed deprecated code