Qualify every use of Prelude functions (#131)

[sgraf812]

This bit end users in visq/language-c#99.

Properly fixes #131 and adds a regression test.

[Bodigrim]

This is a great improvement, but it might be nice to document somewhere that happy relies on Prelude being base:Prelude, because in general it can be overridden in Cabal file with something like mixins: base hiding (Prelude), relude (Relude as Prelude).

(I'm not sure whether it's possible or practical to resolve the general issue)

[sgraf812]

I had hoped to import Prelude qualified as Happy_Pre and then use a package import, but that means that the implicit import Prelude is no longer inserted (Section 5.6.1 of the Haskell report).

So happy cannot by itself add a qualified import of Prelude without surprising effect on user programs.
It appears that users who import Prelude (...) or redefine Prelude must also import qualified "base" Prelude. It's a bit ugly, but not too complicated to detect and adhere to.

I documented this in the user's guide.

[andreasabel]

This PR broke Agda:

• Happy-2.1.1 causes Agda build to fail agda/agda#7585

Because our parser has import Prelude hiding (null) another import qualified Prelude has to be omitted by the generator to have Prelude.null in scope.

I think 2.1.1 should be deprecated as it pretends to be a minor bump but breaks user's code.

I'll open a new issue for all that, no need to reply here.