[turbofan] Remove the FloatXXSubPreserveNan operators.

src/compiler/instruction-selector.cc

Index: src/compiler/instruction-selector.cc
diff --git a/src/compiler/instruction-selector.cc b/src/compiler/instruction-selector.cc
index 2cee88e41fde5d51fc2da25ec8bbac82df44c743..e19de56811299616449bc78f8cf5b620db8aaae0 100644
--- a/src/compiler/instruction-selector.cc
+++ b/src/compiler/instruction-selector.cc
@@ -1134,8 +1134,6 @@ void InstructionSelector::VisitNode(Node* node) {
       return MarkAsFloat32(node), VisitFloat32Add(node);
     case IrOpcode::kFloat32Sub:
       return MarkAsFloat32(node), VisitFloat32Sub(node);
-    case IrOpcode::kFloat32SubPreserveNan:
-      return MarkAsFloat32(node), VisitFloat32SubPreserveNan(node);
     case IrOpcode::kFloat32Neg:
       return MarkAsFloat32(node), VisitFloat32Neg(node);
     case IrOpcode::kFloat32Mul:
@@ -1156,8 +1154,6 @@ void InstructionSelector::VisitNode(Node* node) {
       return MarkAsFloat64(node), VisitFloat64Add(node);
     case IrOpcode::kFloat64Sub:
       return MarkAsFloat64(node), VisitFloat64Sub(node);
-    case IrOpcode::kFloat64SubPreserveNan:
-      return MarkAsFloat64(node), VisitFloat64SubPreserveNan(node);
     case IrOpcode::kFloat64Neg:
       return MarkAsFloat64(node), VisitFloat64Neg(node);
     case IrOpcode::kFloat64Mul: