Fix some typos (found by codespell) (#160)

Signed-off-by: Stefan Weil <sw@weilnetz.de>

bench/icu.c

@@ -41,7 +41,7 @@ int main(int argc, char **argv)
 
 		  /* ICU's insane normalization API requires you to
 			 know the size of the destination buffer in advance,
-			 or alternatively to repeatly try normalizing and
+			 or alternatively to repeatedly try normalizing and
 			 double the buffer size until it succeeds.  Here, I just
 			 allocate a huge destination buffer to avoid the issue. */
 		  UChar *udest = (UChar*) malloc(10*ulen * sizeof(UChar));

bench/unistring.c

@@ -1,4 +1,4 @@
-/* comparitive benchmark of GNU libunistring */
+/* comparative benchmark of GNU libunistring */
 
 #include <stdio.h>
 #include <stdlib.h>

data/charwidths.jl

@@ -52,7 +52,7 @@ const UTF8PROC_CATEGORY_CO = 29
 #############################################################################
 # Use a default width of 1 for all character categories that are
 # letter/symbol/number-like, as well as for unassigned/private-use chars.
-# This can be overriden by UAX 11
+# This can be overridden by UAX 11
 # below, but provides a useful nonzero fallback for new codepoints when
 # a new Unicode version has been released but Unifont hasn't been updated yet.