Spell Check code better

[mcandre]

aspell has some quirks that make spell checking source code awkward:

X) Text on a line after # is unchecked. Y) Long hexadecimal numbers are chopped into (apparently) random blocks of words. Z) String escape characters are partially fused with any words after them.

X makes spell check skip C headers, scripting language comments, and any text after a # in a string. Fortunately, this would be easily solved by #8.

Y would also be easily solved by #8.

Z may take some more work to do correctly, and would certainly require #8.

Adding a basic understanding of C-compatible escape codes like \n could substantially reduce false positives from Z. However, escape codes from .NET, Lisp, and Erlang would still raise their own false positives, proving there's no end to problems as a result of deviating from sprintf.