nir,intel: A couple atomic op reworks
The first commit adds separate signed and unsigned atomic min/max opcodes for images in NIR. This better matches all the other atomic intrinsics such as those for SSBOs and shared variables where the sign is part of the intrinsic opcode. Both generators (GLSL and SPIR-V) know the sign from the type of the image variable or handle. In SPIR-V, signed min/max are separate opcodes from unsigned.
The second commit is to the intel back-end compilers and deletes a pile of switch statements with one that is generated with the aid of a macro. Now that all of the atomic intrinsic opcodes in NIR have consistent naming, it's really easy to macroify the conversion to