[PATCH] [tablegen] Resolve complex def identifiers when instanciating multiclass

[Tim Northover]

> Why are we only calling ResolveMulticlassDefArgs when we can't resolve the name? It seems somewhat funny to have the argument resolution state depend on the kind of name that is used. Does something break if you always call this?

It gets called later on in the caller. Not that I think that improves
things: it looks like a function that's designed to be called
precisely once. Maybe the second call doesn't actually change
anything, but it's horribly ugly.

Cheers.

Tim.