Update docs/advanced_usage.rst

Co-authored-by: Emmett Butler <[email protected]>

Author: danielsn

docs/advanced_usage.rst

@@ -743,7 +743,6 @@ PyTorch Profiling
 
 The PyTorch profiler can be used to trace CPU and GPU events that occur when running inference or training on a PyTorch model.
 The PyTorch profiler as it's `typically used <https://pytorch.org/tutorials/intermediate/tensorboard_profiler_tutorial.html>`__, will output a trace json file to
-
 local disk that can be loaded in a visualization tool like TensorBoard or Perfetto. With the dd-trace-py PyTorch profiler integration, we instrument the `profiler API <https://pytorch.org/docs/stable/_modules/torch/profiler/profiler.html>`__
 to automatically export this data to Datadog for visualization without having to manually copy files between servers.