No test output when running object detection with point pillars

### Checklist

- [X] I have searched for [similar issues](https://github.com/isl-org/Open3D-ML/issues).
- [X] I have tested with the [latest development wheel](http://www.open3d.org/docs/latest/getting_started.html#development-version-pip).
- [X] I have checked the [release documentation](http://www.open3d.org/docs/release/) and the [latest documentation](http://www.open3d.org/docs/latest/) (for `master` branch).

### My Question

Hi, thanks for all the work on this great library! I am following the 'getting started - 3D Object Detection - Running a pretrained model for 3D object detection' part in the Readme.md. I downloaded the KITTI dataset and am using the pointpillars_kitti.yml config file just as in the tutorial. The `pipeline.run_test()` in the end runs without errors:

https://github.com/isl-org/Open3D-ML/blob/538b661c13dc3ffc7737dbfbcc63cbca0e69508b/README.md?plain=1#L228

However, I don't get any results in the newly created test folder. I already traced the problem back to the `save_test_results` method, which always receives an empty list as `attrs` argument for some reason:
https://github.com/isl-org/Open3D-ML/blob/538b661c13dc3ffc7737dbfbcc63cbca0e69508b/ml3d/datasets/kitti.py#L234

Could someone kindly point me in the right direction how to fix this, such that files are actually written in the 'save_test_results' method?