Reduce inspection bonus to close reward farming loophole

cleanops_env/environment.py

@@ -140,7 +140,7 @@ class CleanOpsEnvironment(Environment[DataCleaningAction, DataCleaningObservatio
                 self._focus_table_name = table_name
                 if table_name not in self._state.inspected_tables:
                     self._state.inspected_tables.append(table_name)
-                    insight_bonus = 0.03
+                    insight_bonus = 0.01
                     status_message = f"Inspected table '{table_name}'."
                 else:
                     noop_penalty = -0.02
@@ -155,7 +155,7 @@ class CleanOpsEnvironment(Environment[DataCleaningAction, DataCleaningObservatio
                 self._focus_operation_detail = self._build_operation_detail(self._task_spec, operation_id, self._state.tables, None)
                 if operation_id not in self._state.inspected_operations:
                     self._state.inspected_operations.append(operation_id)
-                    insight_bonus = 0.03
+                    insight_bonus = 0.01
                     status_message = f"Inspected operation '{operation_id}'."
                 else:
                     noop_penalty = -0.02
@@ -342,4 +342,3 @@ class CleanOpsEnvironment(Environment[DataCleaningAction, DataCleaningObservatio
             why_it_matters=operation.why_it_matters,
             change_preview=preview,
         )
-