Similar with other masking support tasks, this task adds the optimization for masking operations for vector "lanewiseShift" API. Basically there is no masked operation for "lanewiseShift", we add it here to make it used for masked binary "lanewise" with shift opcodes and an integer constant as the second argument.