Clarify how string subscripting handles UTF-8.

test/string/literals.wren

@@ -1,2 +1,5 @@
 IO.print("".count)   // expect: 0
 IO.print("a string") // expect: a string
+
+// Non-ASCII.
+IO.print("A~¶Þॐஃ") // expect: A~¶Þॐஃ

test/string/subscript.wren

@@ -10,5 +10,26 @@ IO.print("abcd"[-3]) // expect: b
 IO.print("abcd"[-2]) // expect: c
 IO.print("abcd"[-1]) // expect: d
 
-// Make sure the string's internal buffer size is correct.
-IO.print("abcd"[1] == "b") // expect: true
+// Regression: Make sure the string's internal buffer size is correct.
+IO.print("abcd"[1] == "b") // expect: true
+
+// Indexes by byte, not code point.
+//
+// Bytes:           11111
+//        012345678901234
+// Chars: sø mé ஃ  thî ng
+IO.print("søméஃthîng"[0]) // expect: s
+IO.print("søméஃthîng"[1]) // expect: ø
+IO.print("søméஃthîng"[3]) // expect: m
+IO.print("søméஃthîng"[6]) // expect: ஃ
+IO.print("søméஃthîng"[10]) // expect: h
+IO.print("søméஃthîng"[-1]) // expect: g
+IO.print("søméஃthîng"[-2]) // expect: n
+IO.print("søméஃthîng"[-4]) // expect: î
+
+// If the subscript is in the middle of a UTF-8 sequence, yield an empty string.
+IO.print("søméஃthîng"[2] == "") // expect: true
+IO.print("søméஃthîng"[7] == "") // expect: true
+IO.print("søméஃthîng"[8] == "") // expect: true
+IO.print("søméஃ"[-1] == "") // expect: true
+IO.print("søméஃ"[-2] == "") // expect: true

test/string/utf_8_in_literal.wren

@@ -1,3 +0,0 @@
-IO.print("A~¶Þॐஃ") // expect: A~¶Þॐஃ
-
-// TODO: Malformed UTF-8 source files.