Add new SPI (GWTau) and benchmark generation script + workflow.

[jmoo2880]

In this PR, I've made the following changes:

• Added a new manually executable Git Actions workflow for generating benchmark tables for SPI testing called run_datset_generation.yaml. Upon running (via Git Actions portal), the CML7.npy dataset will be loaded and SPIs will be computed on the same dataset for a fixed number of iterations (using a new Calculator each time). The mean and standard deviation for each SPI's output matrix (MPI) is then saved as a dictionary and uploaded as a .pkl file that can be downloaded as an artefact in Git Actions. Each time a new SPI is added, or a major change (e.g., dependency update) is made, the workflow can be manually triggered to create a new baseline set of SPI outputs.
• Added a new pairwise distance measure - Gromov-Wasserstain (GWTau) distance. See here for more details about the specific implementation.
• Added a new in-depth tutorial for integrating SPIs into the pyspi framework using GWTau as an example. See the docs for a full explanation of the validation/testing process of this SPI using the dataset provided in the original research paper.
• Added new benchmark dataset tables which include reference outputs for the new GWTau SPI.
• Updated the default (config.yaml) and fast (fast.yaml) subset configuration files to include GWTau.
• Updated SPI output testing.