feat: ci, web output, reorganization

[tristan-f-r]

(TLDR): This primarily integrates CI, which runs the yeast-osmotic-stress, hiv, and responsenet datasets, displaying them on a little webpage.

• See the repository: https://github.com/tristan-f-r/spras-benchmarking/
• See the web output: https://tristan-f-r.github.io/spras-benchmarking/

This supersedes #5 and #11 (and, consequently, #7 and #6) by integrating the RN data and the synthetic pathway data into this pull request (for testing the PRA workflow). Note: This PR contains scripts (including from #5 and #11) that process biological data and should be reviewed.

Organization

This closes #21, with the following organization:

.
├── configs
├── datasets
├── spras
└── web

• configs is our SPRAS configs, split by data type (DMMM, PRA)
• datasets is each category of datasets, and is mainly meant for heuristic organization (e.g. hiv and yeast-osmotic-stress are still separate, but synthetic networks that come from the same database would be categorized in the same dataset folder. This isn't an all-encompassing standard and should be broken when necessary)
• spras as a submodule (it is easier to make commits to spras-benchmarking than it is to spras)
• web for web output

Eventually, there will be an analysis folder for further analysis not yet supported by SPRAS, making SPRAS-benchmarking double-serve as a staging area for analysis.

[annaritz]

There is a lot here - I don't have the technical chops to do a full code review without talking through the PR with @tristan-f-r and @AMINOexe. I made a few comments, and look forward to talking more.

One question for @agitter is whether we should perform this merge as a first-attempt at the repo (since the repo is so new), knowing that not all details have been thought out yet. It would help us get a sense of how to run multiple datasets, and would be helpful to have this merged when we think about file storage solutions.

[ntalluri]

The synthetic pathways dataset and Gabes ResponseNet dataset still need a review over the code in general. I think we should open PRs again to review those once this is merged rather than trying to review those in this PR.

[tristan-f-r]

[Waited for CI to pass] I've removed the unreviewed datasets and am splitting it to a separate PR.