fix trainer c and soccer config

config/ppo/SoccerTwos.yaml

     network_settings:
       normalize: false
       hidden_units: 512
-      num_layers: 2
+      num_layers: 3
       vis_encode_type: simple
     reward_signals:
       extrinsic:
-    max_steps: 50000000
+    max_steps: 100000000
     time_horizon: 1000
     summary_freq: 10000
     threaded: false
       window: 10
       play_against_latest_model_ratio: 0.5
       initial_elo: 1200.0
+environment_parameters:
+  ball_touch:
+    curriculum:
+      - name: Lesson0
+        completion_criteria:
+          measure: progress
+          behavior: SoccerTwos
+          signal_smoothing: true
+          min_lesson_length: 100
+          threshold: 0.0005
+        value: 1.0
+      - name: Lesson1
+        completion_criteria:
+          measure: progress
+          behavior: SoccerTwos
+          signal_smoothing: true
+          min_lesson_length: 100
+          threshold: 0.001
+        value: 0.5
+      - name: Lesson2
+        completion_criteria:
+          measure: progress
+          behavior: SoccerTwos
+          signal_smoothing: true
+          min_lesson_length: 100
+          threshold: 0.0015
+        value: 0.1
+      - name: Lesson3
+        completion_criteria:
+          measure: progress
+          behavior: SoccerTwos
+          signal_smoothing: true
+          min_lesson_length: 100
+          threshold: 0.002
+        value: 0.0
+env_settings:
+  num_envs: 2

ml-agents/mlagents/trainers/trainer_controller.py

     def reset_env_if_ready(self, env: EnvManager) -> None:
         # Get the sizes of the reward buffers.
         reward_buff = {k: list(t.reward_buffer) for (k, t) in self.trainers.items()}
-        curr_step = {k: int(t.step) for (k, t) in self.trainers.items()}
+        curr_step = {k: int(t.get_step) for (k, t) in self.trainers.items()}
         max_step = {k: int(t.get_max_steps) for (k, t) in self.trainers.items()}
         # Attempt to increment the lessons of the brains who
         # were ready.