radeonsi: properly implement texture opcodes that take an offset

radeonsi: properly implement texture opcodes that take an offset

Instead of using intr_name in lp_build_tgsi_action, this selects the names
with a switch statement in the emit function. This allows emitting
llvm.SI.sample for instructions without offsets and llvm.SI.image.sample.*.o
otherwise.

This depends on my LLVM changes.

When LLVM 3.5 is released, I'll switch all texture instructions to the new
intrinsics.