patch applied (ghc): Better handling of node parameter in calling conventions

[John Dias]

>>   * Better handling of node parameter in calling conventions
>>    - Previously, the node was taken as a parameter, then ignored,
>>      for static closures. Goofy. Now, the vestigial node parameters
>>      are gone.
> I presume for a top-level function you still reserve R1, even though nothing 
> is passed in it?  The AutoApply code in the RTS assumes this: regardless of

Yes: R1 is reserved. The idiocy before was that R1 was copied in to 
a dead variable. There was no effect on the correctness of the program, 
and the dead assignments would be dropped later, but it was embarrassing.
-j