Add worm config for SAC (#3879)

config/sac_trainer_config.yaml

     max_steps: 5.0e5
     summary_freq: 2000
     time_horizon: 10
+
+WormStatic:
+    normalize: true
+    time_horizon: 1000
+    batch_size: 128
+    steps_per_update: 20
+    buffer_size: 500000
+    buffer_init_steps: 2000
+    max_steps: 3e6
+    summary_freq: 30000
+    init_entcoef: 1.0
+    num_layers: 2
+    hidden_units: 256
+    reward_signals:
+        extrinsic:
+            strength: 1.0
+            gamma: 0.995
+
+WormDynamic:
+    normalize: true
+    time_horizon: 1000
+    batch_size: 128
+    buffer_size: 500000
+    summary_freq: 30000
+    steps_per_update: 20
+    num_layers: 2
+    max_steps: 5e6
+    hidden_units: 256
+    reward_signals:
+        extrinsic:
+            strength: 1.0
+            gamma: 0.995