Classification without fix_intercept now included in all files

1) Width reduced to 1e-5
2) SVM done with a C=100
3) SVM done with Kernel SVM but with a linear Kernel
4) Fixes to how algorithm breaks ties
5) Added a python SVM function
6) Modified Model python interface to accomadate SVM functions

use a linear prgramming method to check the feasabilty of a convex
combination of the classes to get an estimate of the number of points
in overlapping convex hull.

Compiles but not tested

Create a base class for DI and setup for Classifier

Projection score added for classification problems, now onto the di part

Feature selected on a particular rank but not overall were
not set selected to false

fix some bugs from previous merge

Don't always reuse

n_models_store now sets how many models to store (default still n_residual)

For the role descriptor

can set a maximum cross correlation for selected features

Properties now have units and postprocessing scripts avaiable

minor tweaks to output of Models and fixed a memory bug

basic analysis scripts completed

Using these expressions one can reconstruct features using the primary feature set
Capability to decompse features into the prevelance of primary features in them

Use keyed files to automatically generate bindings.hpp/bindings.cpp