use safe negative binomial rng, take 2

[sbfnk]

Description

This PR closes #963 again (looks like a merge fail in 4b46352 removed the fix).

Initial submission checklist

• My PR is based on a package issue and I have explicitly linked it.
• I have tested my changes locally (using devtools::test() and devtools::check()).
• I have added or updated unit tests where necessary.
• I have updated the documentation if required and rebuilt docs if yes (using devtools::document()).
• I have followed the established coding standards (and checked using lintr::lint_package()).
• I have added a news item linked to this PR.

After the initial Pull Request

• I have reviewed Checks for this PR and addressed any issues as far as I am able.

[github-actions]

Thank you for your contribution sbfnk 🚀! Your synthetic_recovery markdown is ready for download 👉 here 👈!
_{(The artifact expires on 2025-02-09T13:20:06Z. You can re-generate it by re-running the workflow here.)}

[github-actions]

This is how benchmark results would change (along with a 95% confidence interval in relative change) if 4475677 is merged into main:

• ✔️default: 23.4s -> 21.8s [-15.54%, +2.1%]
• ✔️no_delays: 25.8s -> 24.2s [-24.96%, +12.71%]
• ✔️random_walk: 10.4s -> 10.7s [-24.33%, +30.74%]
• ✔️stationary: 12.1s -> 12.4s [-8.33%, +12.92%]
• ✔️uncertain: 34.4s -> 36.9s [-2.69%, +17.37%]
  Further explanation regarding interpretation and methodology can be found in the documentation.

[github-actions]

This is how benchmark results would change (along with a 95% confidence interval in relative change) if 285aac8 is merged into main:

• ✔️default: 22.1s -> 23.4s [-6.76%, +18.09%]
• ✔️no_delays: 31.5s -> 23.8s [-58.59%, +10.11%]
• ✔️random_walk: 9.49s -> 9.08s [-12.35%, +3.71%]
• ✔️stationary: 12.7s -> 12.5s [-10.74%, +7.73%]
• ✔️uncertain: 33.4s -> 36.4s [-0.05%, +17.79%]
  Further explanation regarding interpretation and methodology can be found in the documentation.