`PUPPYPRINT_DEBUG` crashes upon loading into a real course (3.0 only)

[gheskett]

Seems like maybe some sort of memory-related conflict was introduced here? This is far from an ideal crash message.

Interestingly, the castle grounds audio is still playing post-crash.

[Anoliah]

Well I've managed to track down the cause of this bug. #796 updated some of the segment ids and the puppyprint debug segment profiler was not updated to account for this.
The problematic function seems to be set_segment_memory_printout, which ends up indexing ramsizeSegment out of bounds due to the segment id changes.
That being said I don't know what the best way to fix this is because I have absolutely no idea why the puppyprint debug code was written like this to begin with...

[gheskett]

Issue addressed. After having fixed it...I also don't really understand why that page was written like this.