Fix lr schedule

Call lr scheduler before training so that the update takes place properly also if a device missed out one or several training rounds. This happends e.g. when changing the server device in Swarm SL or neglecting client updates if using a loss threshold. Also added lr scheduler for client for parallel SL.

edml/controllers/test_controller.py

@@ -43,7 +43,7 @@ class TestController(BaseController):
 
     def _get_model_with_highest_postfix_number(self, model_save_path):
         """Returns the highest postfix number in the given directory for the configured model_prefix."""
-        model_prefix = self.__model_prefix__()
+        model_prefix = f"{self.__model_prefix__()}_client_"  # assume server and client weights were saved appropriately
         highest_postfix_number = 0
         for file in os.listdir(model_save_path):
             if file.startswith(model_prefix):