Git scrape Product and Group information for all agencies

[thekaveman]

We would like to understand changes over time to the Littlepay Product and Group details for each agency we work with. This has come up recently as we learned about a change to a Product ID after that Product was duplicated to create a new Product -- we didn't expect this to create a TWO new Products 🤷

So we'd like to capture the output of our littlepay groups and littlepay products CLI commands, in CSV files stored in this repository, to be able to assess those changes over time more easily.

Acceptance Criteria

• For each agency in production (MST, SacRT, SBMTD as of opening this issue)
• On some schedule (at least once per day)
• Run littlepay groups and littlepay products
• Store the data in a corresponding CSV file in this repository: data/[agency slug]_[groups|products].csv
• With column headers matching the fields in the corresponding GroupResponse and ProductResponse objects
• Also store the output of littlepay groups products in a CSV file in this repository: data/[agency slug]_linked_groups_products.csv
• With column headers group_id, product_id
• A new commit is made each time this process runs, for each agency, when any data file is changed

Additional context

Sort of related to the work in cal-itp/benefits#1889, can be implemented in workflow in a similar fashion.

More about the git scraping technique.

[thekaveman]

cc @indexing @o-ram, here's the ticket write up for collecting Littlepay Group and Product info over time in this littlepay repository.

Question for you both: how frequently and when should this process run?

[thekaveman]

Going with once per day in the evening (roughly 10-11pm Pacific) for now -- this will capture any changes made throughout the day, and we'll note the timestamp of when we capture the changes to hopefully make later analysis a little easier.

[indexing]

@thekaveman The time and frequency sound good to me; seems like a good match for the infrequnecy with which products change.

[o-ram]

@thekaveman also sounds good to me!