to using MD5 hashes to identify TypeReps in the Typeable library.

to the Typeable library.  We now generate an MD5 hash of the
fully-qualified TyCon name at compile time.

No concrete syntax yet, but I need to be able to use these in code
generated for derived Typeable instances.

The seq# case in the new codegen was being shadowed by a more general
case.

Signed-off-by: Edward Z. Yang <ezyang@mit.edu>

This is safe because GHC never generates a fast call to a data constructor
worker: if the call is seen statically it will be eta-expanded and the
allocation of the data will be inlined. We still need to export the _closure
in case the constructor is used in an unapplied fashion.

I introduced this to support explicitly recording the info table label
in RawCmm for another patch I am working on, but it turned out to lead
to significant simplification in those parts of the compiler that
consume RawCmm.

Now, instead of lots of tests for null [CmmStatic] we have a simple
test of a Maybe, and have reduced the number of guys that need to know
how to convert entry->info labels by a TON. There are only 3 callers
of that function now!

Signed-off-by: George Roldugin <groldugin@cse.unsw.edu.au>

I observed that the [CmmStatics] within CmmData uses the list in a very stylised way.
The first item in the list is almost invariably a CmmDataLabel. Many parts of the
compiler pattern match on this list and fail if this is not true.

This patch makes the invariant explicit by introducing a structured type CmmStatics
that holds the label and the list of remaining [CmmStatic].

There is one wrinkle: the x86 backend sometimes wants to output an alignment directive just
before the label. However, this can be easily fixed up by parameterising the native codegen
over the type of CmmStatics (though the GenCmmTop parameterisation) and using a pair
(Alignment, CmmStatics) there instead.

As a result, I think we will be able to remove CmmAlign and CmmDataLabel from the CmmStatic
data type, thus nuking a lot of code and failing pattern matches. This change will come as part
of my next patch.

It was already deprecated, and marked for removal in 7.2.

We no longer advertise that we support the Generics extension,
although we still do accept -XGenerics and -XNoGenerics flags
(but warn that we are ignoring them).

I also remove the even older -fgenerics and -fno-generics flags.

It used to take a whole DynFlags, but that meant we had to
create a DynFlags with (panic "No settings") for settings, as
we didn't have any real settings.

Now we just pass the LogAction, which is all that it actually needed.
The default is exported from DynFlags as defaultLogAction.

(fixes various ffi test failures on x86.  This change was supposed to
be part of 9f61598c, but somehow it
got lost).