machinst lowering: update inst color when scanning across branch to allow more load-op merging.

A branch is considered side-effecting and so updates the instruction
color (which is our way of computing how far instructions can sink).
However, in the lowering loop, we did not update current instruction
color when scanning backward across branches, which are side-effecting.
As a result, the color was stale and fewer load-op merges were permitted
than are actually possible.

Note that this would not have resulted in any correctness issues, as the
stale color is too high (so no merges are permitted that should have
been disallowed).

Fixes #2562.

cranelift/filetests/filetests/isa/x64/load-op.clif

@@ -59,3 +59,14 @@ block0(v0: i64, v1: i64):
   ; nextln: movq    0(%rax,%rdi,1), %rsi
   ; nextln: movq    %rsi, %rax
 }
+
+function %merge_scalar_to_vector(i64) -> i32x4 {
+block0(v0: i64):
+  v1 = load.i32 v0
+  v2 = scalar_to_vector.i32x4 v1
+  ; check: movss   0(%rdi), %xmm0
+
+  jump block1
+block1:
+  return v2
+}