Add "feature distributions" experiment, other changes

[mrcslws]

Various changes:

• Add support for various object distributions. Move this code into a shared object_generation.py.
• New experiment and plot: feature distributions
• New script that generates the heatmap using the capacity simulation results rather than the convergence results.
• Converted some .sh scripts to call a python script which prepares a set of experiments and then calls runExperiments rather than running a series of convergence_experiment.py command. This enables us to queue all of the work at once, so the script effectively uses all available CPUs and finishes faster.
• Move the gaussian_simulation and its associated scripts to a "entorhinal" subfolder that focuses on modeling parameters like "readout resolution".
• Enable using the gaussian model from convergence_simulation and capacity_simulation. Rather than exposing full access to the gaussian parameters via this script, retrofit the gaussian model to use these parameters so that it plugs nicely into our existing plotting scripts.
• Fix bug in createRatModule that I added during code cleanup. It needs to use the enlargeModuleFactor to scale down the bump.