Spurious example code in docs

[jmackie]

For example: Data.ByteString.split has the type: Word8 -> ByteString -> [ByteString], yet the code examples suggest it takes a Char.

-- > split '\n' "a\nb\nd\ne" == ["a","b","d","e"]
-- > split 'a'  "aXaXaXa"    == ["","X","X","X",""]
-- > split 'x'  "x"          == ["",""]

I suspect this might be deliberate as the samples are instructive, and in the above example not using Char would be much less clear, but nonetheless this could be a source of confusion for newbies. Surely all the documentation code should compile? I'll happily open a PR to integrate doctest for this?