docs: clarify string is byte array, move .count to codePoints.count, remove O(n) primitive

The string class is now documented as an immutable byte array rather than a
sequence of Unicode code points. The .count getter is removed from the string
primitive and replaced by .codePoints.count on the code point sequence
interface, making the O(n) cost explicit. Updated the doc site and test files
accordingly.

test/core/string/count.wren

@@ -6,3 +6,13 @@ IO.print("\0".count)  // expect: 1
 IO.print("a\0b".count)  // expect: 3
 IO.print("\0c".count)  // expect: 2
 IO.print(("a\0b" + "\0c").count)  // expect: 5
+
+// Treats a UTF-8 sequence as a single item.
+//
+// Bytes:           11111
+//        012345678901234
+// Chars: sø mé ஃ  thî ng
+IO.print("søméஃthîng".count) // expect: 10
+
+// Counts invalid UTF-8 one byte at a time.
+IO.print("\xefok\xf7".count) // expect: 4