Update sharded_moe.py for formatting

Author: xenshinu

deepspeed/moe/sharded_moe.py

@@ -535,7 +535,8 @@ def forward(self,
 
         elif self.k == 2:
             gate_output = top2gating(logits, self.capacity_factor if self.training else self.eval_capacity_factor,
-                                     self.min_capacity, self.drop_tokens, self.ep_group, self.top2_2nd_expert_sampling, use_tutel)
+                                     self.min_capacity, self.drop_tokens, self.ep_group, self.top2_2nd_expert_sampling, 
+                                     use_tutel)
         else:
             gate_output = topkgating(logits, self.k,
                                      self.capacity_factor if self.training else self.eval_capacity_factor,