Testsuite failures on several architectures

As part of the automatic building of bornagain on several architectures by the Debian project, several testsuite failures have been discovered. Most of them (from what I can see currently) are due to calculation errors, with the difference between expected and calculated value being over the acceptable threshold. Sometimes by a small amount, but sometimes by as much as a 10⁹ factor: I see 393: FAILED: relative deviation of dat from ref is 0.181818, above given threshold 2e-10 for arm64 for instance.

The build logs for the current version of bornagain, including the reported errors, is available on https://buildd.debian.org/status/package.php?p=bornagain (you'll want to have a look at the "Build-Attempted" lines).

I can do manual builds on porterboxes if it helps testing patches.