fix(cpu): PPCBUG-426/427/433 single-FMA vnmsubfp + vctsxs NaN saturation

Phase 5 batch 6 (5f): saturation and FMA-rounding fixes.

- PPCBUG-426 vnmsubfp: was `bi - ai * ci` (two rounding steps); now
  `-ai.mul_add(ci, -bi)` which is mathematically equivalent (= bi - ai*ci)
  but uses a single FMA round per ISA.
- PPCBUG-427 vnmsubfp128: same single-FMA fix.
- PPCBUG-433 vctsxs / vcfpsxws128 NaN saturation: AltiVec ISA saturates
  NaN to INT_MIN (0x80000000); xenia returned 0. The vctuxs (unsigned)
  NaN→0 is correct per ISA.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

crates/xenia-cpu/src/interpreter.rs

@@ -2039,7 +2039,8 @@ fn execute(ctx: &mut PpcContext, mem: &dyn MemoryAccess, instr: &DecodedInstr) -
                 let ai = vmx::flush_denorm(a[i]);
                 let bi = vmx::flush_denorm(b[i]);
                 let ci = vmx::flush_denorm(c[i]);
-                r[i] = vmx::flush_denorm(bi - ai * ci);
+                // PPCBUG-426: single FMA rounding instead of two-step (b - a*c).
+                r[i] = vmx::flush_denorm(-ai.mul_add(ci, -bi));
             }
             ctx.vr[instr.rd()] = xenia_types::Vec128::from_f32x4_array(r);
             ctx.pc += 4;
@@ -2056,7 +2057,8 @@ fn execute(ctx: &mut PpcContext, mem: &dyn MemoryAccess, instr: &DecodedInstr) -
                 let ai = vmx::flush_denorm(a[i]);
                 let bi = vmx::flush_denorm(b[i]);
                 let di = vmx::flush_denorm(d[i]);
-                r[i] = vmx::flush_denorm(di - ai * bi);
+                // PPCBUG-427: single FMA rounding.
+                r[i] = vmx::flush_denorm(-ai.mul_add(bi, -di));
             }
             ctx.vr[instr.vd128()] = xenia_types::Vec128::from_f32x4_array(r);
             ctx.pc += 4;