Assess if we want to develop a data versioning strategy for raw indices data

[jdhoffa]

And also potentially for processed indices data.

Relates to #21 (comment)

AB#9893

[jdhoffa]

Relevant comment:

This looks good to me.
However, there is one consideration we should make, and that is if we want this to run the scraping every time we run this workflow.

This workflow achieves three main steps

Scraping the raw indices
Preparing those indices as PACTA input
Running PACTA on them
I can envision a situation where we may want to only run the last step (for example, if we fix a bug in PACTA. We may want to update the indices output just to include the bugfix, but not necessarily be fussed about pulling the indices fresh again).

What do you think?