Use ghc return value instead of looking for the stderr, do not throw error when an haskell module throws a build failure, e.g. because of false positives like this one (debian/arm*) "You are using a new version of LLVM that hasn't been tested yet! We will try though..."

Was not a good idea in the first place IMO

Workaround bug (each time losing room master status, even when joining mutliple rooms, new instance of NetAmmoSchemeModel created, receiving schemeConfig and modifying its 43rd member, thus the last model which accepts this signal has the string cut down several times, workaround creates copy of qstringlist to avoid modifying shared message instance. Proper fix would delete unneeded instances of NetAmmoSchemeModel, but who cares)