Invariant theory: document whether our conventions match (or not) Sturmfels, Derksen/Kemper, Disquisitiones Arithmetica, Magma, ...

[fingolfin]

See here for some context: #3443 (comment)

In a nutshell, there are various choices (such as whether to use left or right actions and where) to be made, that can lead to different results when computing invariants of certain groups. This can be surprising for users who may not be aware of this. So we should briefly discuss it in the manual, possibly also together with a short example that highlights the difference. We have at least one example in the book by Sturmfels vs. Derksen/Kemper, IIRC, where this became clear (@Lax202 probably remember and can provide the exact "coordinates" / page numbers in either book)