[arm64] Implement Float(32|64)(Min|Max) using fcsel.

src/compiler/machine-operator.h

Index: src/compiler/machine-operator.h
diff --git a/src/compiler/machine-operator.h b/src/compiler/machine-operator.h
index d2c80573020dff6e26a3d8b022a9a63a8ed4cf75..2a80feecded6f01574f7c156228af6e41573c115 100644
--- a/src/compiler/machine-operator.h
+++ b/src/compiler/machine-operator.h
@@ -105,7 +105,7 @@ class MachineOperatorBuilder final : public ZoneObject {
   // for operations that are unsupported by some back-ends.
   enum Flag {
     kNoFlags = 0u,
-    // Note that Float*Max behaves like `(a < b) ? b : a`, not like Math.max().
+    // Note that Float*Max behaves like `(b < a) ? a : b`, not like Math.max().
     // Note that Float*Min behaves like `(a < b) ? a : b`, not like Math.min().
     kFloat32Max = 1u << 0,
     kFloat32Min = 1u << 1,