Fix some i128 shift-related bugs in x64 backend.

This fixes #2672 and #2679, and also fixes an incorrect instruction
emission (`test` with small immediate) that we had missed earlier.

The shift-related fixes have to do with (i) shifts by 0 bits, as a
special case that must be handled; and (ii) shifts by a 128-bit amount,
which we can handle by just dropping the upper half (we only use 3--7
bits of shift amount).

This adjusts the lowerings appropriately, and also adds run-tests to
ensure that the lowerings actually execute correctly (previously we only
had compile-tests with golden lowerings; I'd like to correct this for
more ops eventually, adding run-tests beyond what the Wasm spec and
frontend covers).

cranelift/codegen/src/isa/x64/inst/emit.rs

@@ -1371,7 +1371,7 @@ pub(crate) fn emit(
                 RegMemImm::Imm { simm32 } => {
                     // FIXME JRS 2020Feb11: there are shorter encodings for
                     // cmp $imm, rax/eax/ax/al.
-                    let use_imm8 = low8_will_sign_extend_to_32(*simm32);
+                    let use_imm8 = is_cmp && low8_will_sign_extend_to_32(*simm32);
 
                     // And also here we use the "normal" G-E ordering.
                     let opcode = if is_cmp {