[JOSS-Review] Paper suggestions

[jonaspleyer]

Here is a unstructured list of suggestions to improve the paper:

• Add LAPACK citation (line 29)
• Add ndarray citation (Table 1; Description)
• Briefly discuss possible known downsides of the matrix memory layout as Vec<f64> as opposed to other approaches. I believe there were some blog posts when ndarray was first created about this topic (maybe rather nalgebra I am not completely certain)
• Include enzyme in the discussion. How does this compare to the AD functionality of peroxide? https://github.com/EnzymeAD/Enzyme https://docs.rs/enzyme/latest/enzyme/ https://www.youtube.com/watch?v=FR4fwGsBYPI

I will continue to add more points to this issue as I continue my review.

openjournals/joss-reviews#10366 (comment)