* Add command line argument keep-id, which maintiains randomly generated cell ids. Otherwise cell ids are assigned incrementally (after the removal of cells), which should keep them consistent across runs in version control

* Modify test_cell and test_exception in test_keep_output_tags.py to use the new strip_output signature

* Fix failed test_end_to_end_nbstripout with test_max_size by passing --keep-id for keeping the existing ids

* Add tests for notebooks with and without the --keep-id flag. A new extension expected_id was added for expected output with ordered ids

* Modify the readme to include the --include-id flag

* Add keyword arguments for None inputs in test_keep_output_tags.py

* Rename expected output files to make desired sequential ids more explicit

Co-authored-by: Florian Rathgeber <florian.rathgeber@gmail.com>