Cranelift: x64 backend is missing general lowerings for uunarrow

[elliottt]

.clif Test Case

function %f1(i64x2, i64x2) -> i32x4 {                                                       
block0(v0: i64x2, v1: i64x2):                                                                   
  v2 = uunarrow v0, v1                                                        
  return v2                                                                          
}  

Steps to Reproduce

Compile the example clif, targeting x86_64.

Expected Results

Code would be generated.

Actual Results

A panic is generated, because we currently only support lowering this operation in the context of a fcvt_to_uint_sat and zero input: the following program will compile:

function %f1(f64x2) -> i32x4 {                                                       
block0(v0: f64x2):                                                                   
  v1 = fcvt_to_uint_sat.i64x2 v0                                                     
  v2 = vconst.i64x2 [0 0]                                                            
  v3 = uunarrow v1, v2                                                               
  return v3                                                                          
}      

https://github.com/bytecodealliance/wasmtime/blob/ca6d648e37c80cb13e5027cc746ec0d2e4bd64d1/cranelift/codegen/src/isa/x64/lower.rs#L619

Versions and Environment

Cranelift version or commit: ca6d648e37c80cb13e5027cc746ec0d2e4bd64d1

Operating system: Linux

Architecture: x86_64

Extra Info

This is a similar issue to #4734.