radeonsi: fix compute blit shader compilation when use aco

only vectorize nir ops that aco support

To fix si_compute_blit created nir code compilation with ACO. Two 16bit vector ops are used in it:

      con 16x2  %11 = u2u16 %10.xy
      con 16x2  %25 = f2f16 %22.xy

which is not supported by ACO yet.