[Open data] Streamline cleanup for resource types open data

⚠️ Problem

Building open data generates both a CSV and a JSON file for resource types. Before committing those, we need to review and remove any undesirable changes. Currently, the CSV and JSON files are built at the same time, so this manual step has to happen twice. We should change that to only happen once, then build the other file after the first was confirmed. Either that or deprecate one of the files.

🛠️ Solution

Maybe standardize on CSV files in Build-OpenData and then convert to JSON in the Package-Toolkit script, where we get ready for the release.

The final design is open. The goal is to streamline the update process.

🙋‍♀️ Ask for the community

We could use your help:

1. Please vote this issue up (👍) to prioritize it.
2. Should we limit the release to CSV, JSON, or another format?
3. Do you feel we must support multiple formats?
4. Leave comments to help us solidify the vision.

Read the Build-OpenData and Package-Toolkit entry points to map where the resource-type CSV and JSON files are generated and prepared for release. First clarify whether CSV, JSON, or another format is the supported source of truth, then define a single review step and verify that the remaining format is produced afterward.