Create better (and simpler) plotting routines for density based approach

[eecsu]

2d marginal scatter plots with r values (using get_weights from the sample set) should be made as well as kde approximations in 1 and 2d based on r values.

This should be in a different module since plotP.py was made for plotting based on probability computations not densities. Suggested name of new module: plotDensities.py.

One such 2d scatter plot could be done as simply as follows:

plt.scatter(input_samples.get_values()[:,0], input_samples.get_values()[:,1],
c=input_samples.get_weights())

or something to that effect.