docs: make it clear in the docs that write_delay_ms affects graphql calls

[felixlut]

While trying to make the github_organization datasource faster I noticed that each call to the GraphQL API was heavily throttled, with the provider sleeping exactly 1s in between each call. Doing some digging I've come to the conclusion that the write_delay_ms setting in the provider is the cause, which I've also confirmed by simply changing its value and testing. After some digging it turns out that GraphQL API:s usually use POST requests under the hood. The GitHub GraphQL client is no different.

There probably is a case to be made for separating the settings for the REST and CodeQL API:s, as they are currently both using the value from write_delay_ms. I'll leave that as future work though, as both of these are using the same HTTP client under the hood, and I have no idea if breaking that apart could have some unforeseen consequences. Documenting the behavior is a step forward at least!

---

Before the change?

• No change in functionality, only docs

After the change?

• No change in functionality, only docs

Pull request checklist

• Tests for the changes have been added (for bug fixes / features)
• Docs have been reviewed and added / updated if needed (for bug fixes / features)

Does this introduce a breaking change?

Please see our docs on breaking changes to help!

• Yes
• No

---

[nickfloyd]

Thank you for the research and docs on this ❤️