Add initial VDR generation pipeline

[Scanteianu]

Migration from Scanteianu/openjdk-cve-parser#2, which in turn is a fork of https://github.com/Scanteianu/openjdk-cve-parser.

Stuff in the data directory can be ignored, it's essentially a cache (except for vdr.json, which is the final output)

[smlambert]

Thanks @Scanteianu !

Perhaps remove the cached .json files from the data folder so they do not get checked in with the rest of these changes, if they are just a side-effect of a local run. Can also add a .gitignore file so that when someone is working on PRs for this repository and running tests locally, they do not accidentally include working / cached files in their pull requests.

[Scanteianu]

Thanks @Scanteianu !

Perhaps remove the cached .json files from the data folder so they do not get checked in with the rest of these changes, if they are just a side-effect of a local run. Can also add a .gitignore file so that when someone is working on PRs for this repository and running tests locally, they do not accidentally include working / cached files in their pull requests.

Thanks @smlambert , should i keep the downloads of the website? (the htmls)? In an ideal world i'd put all this stuff in some kind of database, but i'm not sure we need that much complexity yet. The goal is to avoid putting load on nist/ojvg every time we run the tool

[smlambert]

Thanks @Scanteianu !!

Will merge this, then create several issues for next steps (setting up regular run, potentially integrating into overall release pipeline, etc).