diff: Treat multi-byte UTF-8 runes as word characters

Inline diffs on multi-byte UTF-8 characters would match individual
bytes, causing garbled diffs in some cases. For example, replacing
`⊢` with `⊣`, which differ in the final byte only, caused the
diff to display a diff of the bytes instead the character.

This commit uses a workaround present in Mercurial by treating all
bytes 0x80 and above as word characters, causing any multi-byte
character to be treated as a word and not segmented.

https://www.mercurial-scm.org/repo/hg/file/6.3.3/mercurial/patch.py#l51

lib/src/diff.rs

@@ -44,7 +44,12 @@ pub fn find_line_ranges(text: &[u8]) -> Vec<Range<usize>> {
 
 fn is_word_byte(b: u8) -> bool {
     // TODO: Make this configurable (probably higher up in the call stack)
-    matches!(b, b'A'..=b'Z' | b'a'..=b'z' | b'0'..=b'9' | b'_')
+    matches!(
+        b,
+        // Count 0x80..0xff as word bytes so multi-byte UTF-8 chars are
+        // treated as a single unit.
+        b'A'..=b'Z' | b'a'..=b'z' | b'0'..=b'9' | b'_' | b'\x80'..=b'\xff'
+    )
 }
 
 pub fn find_word_ranges(text: &[u8]) -> Vec<Range<usize>> {
@@ -675,6 +680,11 @@ mod tests {
         assert_eq!(find_word_ranges(b"   Abc"), vec![3..6]);
     }
 
+    #[test]
+    fn test_find_word_ranges_multibyte() {
+        assert_eq!(find_word_ranges("⊢".as_bytes()), vec![0..3])
+    }
+
     #[test]
     fn test_find_lcs_empty() {
         let empty: Vec<(usize, usize)> = vec![];