Performance Boosts and Updated Algorithms in NetworkX

The focus of this proposal is to speed up and update the NetworkX (NX) community detection (CD) and subgraph-isomorphism (SI) algorithms. We will also create a roadmap for an improved visualization API within NX. CD algorithms are used in single-cell analysis as well as omics group identification. CD can identify cell type from images, and group proteins or genes by function. Also called clustering algorithms, or graph partitioning, they appear in gene regulatory network analysis. SI detection identifies network motifs which occur more or less often than expected. This helps identify network structures so are used in bioinformatics and general omics analysis, and medical image processing. For CD, we will add functions implementing the Leiden approach and expand our Louvain CD algorithms to include several new approaches. For SI, we will update both symmetry based (ISMAGS) and exhaustive search (VF-series) algorithms. We will complete our partial implementation of the symmetry based ISMAGS library toolset into NetworkX. We will implement VF3 and unify our interface across all VF algorithms. Drawing is important to many users but we need discussion about how to improve that subpackage. So we'll create a roadmap for how to improve NX visualization tools.