[AMDGPU] Add convergent attribute to WWM.

[sheredom]

Summary

Add the convergent attribute to the WWM intrinsic to stop it ever being sunk out of cfg.

[arsenm]

LGTM, though a test wouldn't hurt

llvm/trunk/include/llvm/IR/IntrinsicsAMDGPU.td

 def int_amdgcn_kill : Intrinsic<[], [llvm_i1_ty], []>;
 
 // Copies the active channels of the source value to the destination value,
 // with the guarantee that the source value is computed as if the entire
 // program were executed in Whole Wavefront Mode, i.e. with all channels
 // enabled, with a few exceptions: - Phi nodes with require WWM return an
 // undefined value.
 def int_amdgcn_wwm : Intrinsic<[llvm_any_ty],
-  [LLVMMatchType<0>], [IntrNoMem, IntrSpeculatable]
+  [LLVMMatchType<0>], [IntrNoMem, IntrSpeculatable, IntrConvergent]
 >;
 
 // Given a value, copies it while setting all the inactive lanes to a given
 // value. Note that OpenGL helper lanes are considered active, so if the
 // program ever uses WQM, then the instruction and the first source will be
 // computed in WQM.
 def int_amdgcn_set_inactive :
   Intrinsic<[llvm_anyint_ty],