fix sac shared

ml-agents/mlagents/trainers/policy/torch_policy.py

                 conditional_sigma=self.condition_sigma_on_obs,
                 tanh_squash=tanh_squash,
             )
-            self.shared_critic = False
+            self.shared_critic = True
 
         # Save the m_size needed for export
         self._export_m_size = self.m_size

ml-agents/mlagents/trainers/sac/optimizer_torch.py

         policy_loss = self.sac_policy_loss(log_probs, q1p_out, masks)
         entropy_loss = self.sac_entropy_loss(log_probs, masks)
 
-        total_value_loss = q1_loss + q2_loss + value_loss
+        if self.policy.shared_critic:
+            policy_loss += value_loss
+            total_value_loss = q1_loss + q2_loss
+        else:
+            total_value_loss = q1_loss + q2_loss + value_loss
 
         decay_lr = self.decay_learning_rate.get_value(self.policy.get_current_step())
         ModelUtils.update_learning_rate(self.policy_optimizer, decay_lr)