support for wrapped schedulefree optimizer when using deepspeed (#3266)

* support for wrapped schedulefree optimizer when using deepspeed

* add comment and lint

src/accelerate/optimizer.py

@@ -126,6 +126,13 @@ def train(self):
         """
         if hasattr(self.optimizer, "train") and callable(self.optimizer.train):
             self.optimizer.train()
+        elif (
+            hasattr(self.optimizer, "optimizer")
+            and hasattr(self.optimizer.optimizer, "train")
+            and callable(self.optimizer.optimizer.train)
+        ):
+            # the deepspeed optimizer further wraps the optimizer
+            self.optimizer.optimizer.train()
 
     def eval(self):
         """