patch for -Wno-long-long and early GNAT compilers

[Richard Kenner]

    (There *is* a reason why it would be desirable - to be able to verify
    that GNAT-compiled-by-GNAT and GNAT-compiled-by-different-Ada generate
    the same code for the same input.  People do do this with the C
    compiler occasionally.)

But it's *very* unlikely to happen for Ada unless you restrict yourself
to what can be called a "C subset".  There are many design choices in how
to implement Ada features and each compiler makes different choices.