fix indent bug in test_auto_unit (#507)

Summary:
bypass-github-export-checks

Pull Request resolved: #507

# Context
I'm working to fix torchtnt pyre targets (coming up in next change) and discovered this bug

# This diff
Fix redundant indentation

Reviewed By: JKSenthil

Differential Revision: D48478812

fbshipit-source-id: 1dc2a28a3930331141ee9d96d9f479334eb4256f

tests/framework/test_auto_unit.py

@@ -1095,25 +1095,25 @@ def on_train_step_end(
             # train_step should not be in the timer's recorded_durations because it overlaps with other timings in the AutoUnit's train_step
             tc.assertNotIn("TimingAutoUnit.train_step", recorded_timer_keys)
 
-        def on_eval_step_end(
-            self, state: State, data: Batch, step: int, loss: torch.Tensor, outputs: Any
-        ) -> None:
-            if self.eval_progress.num_steps_completed_in_epoch == 1:
-                tc = unittest.TestCase()
-                recorded_timer_keys = state.timer.recorded_durations.keys()
-                for k in (
-                    "TimingAutoUnit.on_eval_start",
-                    "TimingAutoUnit.on_eval_epoch_start",
-                    "evaluate.iter(dataloader)",
-                    "evaluate.next(data_iter)",
-                    "TimingAutoUnit.move_data_to_device",
-                    "TimingAutoUnit.compute_loss",
-                    "TimingAutoUnit.on_eval_step_end",
-                ):
-                    tc.assertIn(k, recorded_timer_keys)
-
-                # eval_step should not be in the timer's recorded_durations because it overlaps with other timings in the AutoUnit's eval_step
-                tc.assertNotIn("TimingAutoUnit.eval_step", recorded_timer_keys)
+    def on_eval_step_end(
+        self, state: State, data: Batch, step: int, loss: torch.Tensor, outputs: Any
+    ) -> None:
+        if self.eval_progress.num_steps_completed_in_epoch == 1:
+            tc = unittest.TestCase()
+            recorded_timer_keys = state.timer.recorded_durations.keys()
+            for k in (
+                "TimingAutoUnit.on_eval_start",
+                "TimingAutoUnit.on_eval_epoch_start",
+                "evaluate.iter(dataloader)",
+                "evaluate.next(data_iter)",
+                "TimingAutoUnit.move_data_to_device",
+                "TimingAutoUnit.compute_loss",
+                "TimingAutoUnit.on_eval_step_end",
+            ):
+                tc.assertIn(k, recorded_timer_keys)
+
+            # eval_step should not be in the timer's recorded_durations because it overlaps with other timings in the AutoUnit's eval_step
+            tc.assertNotIn("TimingAutoUnit.eval_step", recorded_timer_keys)
 
 
 class TimingAutoPredictUnit(AutoPredictUnit[Batch]):