Hi,
Thanks for submitting this but I do not think that the R&D team would call this a bug. A protective stop is not technically a STOP state of the program. It is paused, so the bit remains in whatever state it was last left in. If you were to resume the program from the protective stop (which you CAN normally do) then the output would continue to pulse.
The other point to make about this is the external program that is monitoring this bit should be looking for a pulsing state. If the bit remains on or off for too long of a period of time then the external program can interpret this how it wants.
IF you think that we somehow changed the behavior of this bit between releases then please let me know.