i#2626: AArch64 v8.0 decode: fix cmode vector shift fields

This fixes the existing cmode3 handling and adds cmode4.

Also fixes:

BIC <Vd>.<T>, #<imm8>{, LSL #<amount>}
ORR <Vd>.<T>, #<imm8>{, LSL #<amount>}

and adds:

MOVI <Vd>.<T>, #<imm8>{, LSL #<amount>}
MOVI <Vd>.<T>, #<imm8>, MSL #<amount>
MVNI <Vd>.<T>, #<imm8>{, LSL #<amount>}
MVNI <Vd>.<T>, #<imm8>, MSL #<amount>

Issue: #2626