case-table functions clobbering extra slots

[Dave Love]

Things like set-case-syntax-pair clobber the extra slots of the case
table.  Is there a good reason for that?  It isn't documented.

It defeats treating cases where there isn't one-to-one correspondence
between upper and lower (which is even mentioned as a possibility in
(elisp)Case Tables).  An example is Greek final sigma, which currently
isn't treated properly.