Lemmatised test example out of date

[ageorgou]

The example for lemmatised text with Arabic translation used in test_successful_lem_existing_lem seems to be out of date, causing that test to fail:

>       assert text == nammu.atfAreaController.edit_area.getText()
E       assert '&X001001 = J...ر تبت، يوم ٢\n' == '&X001001 = JC...ر تبت، يوم ٢\n'
E         Skipping 203 identical leading characters in diff, use -v to show
E         - ht]AV; ūm[day]N; n
E         + ht]AV; ūm[on the day]'PRP; n

We should check how the text is lemmatised (starting from the unlemmatised version) and update the file contents accordingly.