Change the parallelization to be handled by joblib

[nwlandry]

I did the following in this PR:

• I converted all the parallelization scripts (running and file collecting) to be handled by joblib (~10% faster than python multiprocessing)
• Changed all the input arguments to enforce passing by reference.
• I massively re-organized the repo, putting unused scripts in Extra, renaming the scripts to be more descriptive, having a single notebook for generating each figure, and organizing by figures.
• I wrote a script to infer ZKC for different values of $t_{max}$ to generate figure 1c, which is the posterior similarity vs. $t_{max}$ for ZKC.
• I added a script for generating heatmaps for the configuration model.

[review-notebook-app]

Check out this pull request on 

See visual diffs & provide feedback on Jupyter Notebooks.

---

Powered by ReviewNB

[nwlandry]

Once I get new results and the parallelization scripts have been fixed, this will be ready for review.