[commit/Ada] psymbol search failure due to comparison function discrepancy

[Joel Brobecker]

> Is it possible to construct a synthetic test case using the DWARF assembler?
> 
> I ask because I think that, while the patch is fine, having two
> functions which must remain in sync semantically, where one such
> function is only used by Ada and has no test case for some branches, is
> definitely going to cause future regressions.

I agree. Let me see what I can do...

In the long term, I'd really like the psymtab to have the language,
so we can simply use the same routine for both sorting and searching.
ISTR other cases where I really wished I had access to the language
from the psymtab.

-- 
Joel