Fix code style issues with Black

Author: lint-action

bridger/go_explore_phase_1.py

@@ -277,12 +277,15 @@ def update(self, cache_update: StateSamplerCacheUpdate) -> None:
                 self._cache[new_cache_key] = new_cache_entry
 
 
-def clear_illegal_actions(    trajectory: tuple[int], rewards: tuple[float] ) -> tuple[    tuple[int], tuple[float]]:
+def clear_illegal_actions(
+    trajectory: tuple[int], rewards: tuple[float]
+) -> tuple[tuple[int], tuple[float]]:
     filtered = [(t, r) for t, r in zip(trajectory, rewards) if r > -0.101]
     # Unzip the filtered values into separate tuples
     new_trajectory, new_rewards = zip(*filtered) if filtered else ((), ())
     return new_trajectory, new_rewards
-                
+
+
 def rollout(
     rollout_params: RolloutParams,
     start_current_best_trajectory_length: int,
@@ -321,14 +324,18 @@ def rollout(
                     # the validity of the rest of the trajectory
                     # because illegal actions cost reward but do not
                     # change state.
-                    current_trajectory, rewards = clear_illegal_actions(current_trajectory, rewards)
-                    success_entry = SuccessEntry(trajectory=current_trajectory, rewards=rewards)
+                    current_trajectory, rewards = clear_illegal_actions(
+                        current_trajectory, rewards
+                    )
+                    success_entry = SuccessEntry(
+                        trajectory=current_trajectory, rewards=rewards
+                    )
                     if success_entry not in success_entries:
                         success_entries.add(
                             SuccessEntry(trajectory=current_trajectory, rewards=rewards)
                         )
                         led_to_something_new = True
-                        
+
                     state_sampler_cache_update.update_current_best_trajectory(
                         len(current_trajectory)
                     )