Explicit mask vs splat bitcast in masked load for _mm_load_sbh

[RiverDave]

In CIR lowering, we currently emit the mask operand for: _mm_load_sbh (which calls llvm.masked.load) as an explicit vector:

    <8 x i1> <true, false, false, false, false, false, false, false>

whereas the traditional LLVM IR lowering emits:

    <8 x i1> bitcast (<1 x i8> splat (i8 1) to <8 x i1>)

Both are semantically equivalent but should be unified eventually for parity with OG.

related: #1763