Elpi: check if the -j 1 opam patch can be removed about platform

[gares]

I was not aware of the problem. What is it?

[MSoegtropIMC]

We discussed it a while back - elpi does (or did) not compile reliably in CI without -j 1. It failed in roughly 1 out of 4 runs. With -j 1 it does run reliably. I took it as action item to check if this was gone by some magic but I couldn't find time to do this (one should do 20 CI runs to be sure) so I left it in. I will remove it after the release, so that we can have proper CI statistics until 8.17.

[MSoegtropIMC]

It is not a big deal since elpi also compiles with -j 1 reasonably fast and it is not on the critical path, even with -j 12 or so.

[gares]

OK