constant switch test bug in x86Translate

[Stephen Weeks]

> I'm looking into it, but are you reasonable confident that the program at
> this point is valid?

No.  But the C backend did generate a G1 that shows the same problem (not a seg
fault) that I have been fighting with for a few days now.

> Essentially, the error above is arising when I try
> to translate an 
> MachineOutput.Operand.Offset {base, offset, ty}
> but the base turned out to be an immediate value (i.e. a constant int,
> char, word, or label).

With a lot of simplifications turned off, maybe this could happen in some dead
code.  I'll look into it as well.

> I can translate it into something, but it doesn't
> seem to be a reference to a location on the heap. 

Maybe you could generate a call to MLton_bug?  Although it's really nicer to get 
the error earlier, at compile time.  Let's wait until we see what's causing it.