Documentation: Fix last value of index range in FTP listing response

[mgrojo]

• Wording and notation consistent with sf*Shape_{g,s}etPoint()
• Related case fix in sfConvexShape_setPoint

[ChrisThrasher]

I reverted the PR back to its original state since that was better than what I was recommending. Thanks for the fix!