Implement Cluster Analysis

[dakota-hawkins]

• [x] Subset down to most informative genes (variability, tf-idf genes, etc.)
• [x] Transform informative genes to UMAP space
• [x] Perform Louvain clustering on data using UMAP transformation
• [x] Perform pairwise G-Tests between treatments.
• [x] Create plots showing deviation between expected and observed, denote statistically significant differences.