fix: Prevent class selector from doing substring match

Previously, "aa" would have matched "baaa". Now the matching is strict equality
on tokenized words.

src/browser/tests/document/get_elements_by_class_name.html

@@ -15,6 +15,8 @@
   <div class="nested">nested2</div>
 </div>
 
+<div class="baaa">overlap1</div>
+
 <script>
   const foos = document.getElementsByClassName('foo');
 </script>
@@ -96,3 +98,10 @@
   testing.expectEqual('nested1', nested[0].textContent);
   testing.expectEqual('nested2', nested[1].textContent);
 </script>
+
+<script id=overlappingSubstring>
+  // "baaa" is a single class token; "aa" is a substring of it but not its own
+  // token, so it must not match.
+  const overlap = document.getElementsByClassName('aa');
+  testing.expectEqual(0, overlap.length);
+</script>

src/browser/webapi/selector/Selector.zig

@@ -187,37 +187,23 @@ pub fn classAttributeContains(class_attr: []const u8, class_name: []const u8) bo
 }
 
 pub fn classAttributeContainsCase(class_attr: []const u8, class_name: []const u8, case_insensitive: bool) bool {
-    if (class_name.len == 0 or class_name.len > class_attr.len) {
+    if (class_name.len == 0) {
         return false;
     }
 
-    var search = class_attr;
-    while (true) {
-        const pos = (if (case_insensitive)
-            std.ascii.indexOfIgnoreCase(search, class_name)
-        else
-            std.mem.indexOf(u8, search, class_name)) orelse break;
-
-        const is_start = pos == 0 or isClassWhitespace(search[pos - 1]);
-        const end = pos + class_name.len;
-        const is_end = end == search.len or isClassWhitespace(search[end]);
-
-        if (is_start and is_end) return true;
-
-        search = search[pos + 1 ..];
+    var it = std.mem.tokenizeAny(u8, class_attr, &[_]u8{ '\t', '\n', 0x0C, '\r', ' ' });
+    while (it.next()) |token| {
+        if (case_insensitive) {
+            if (std.ascii.eqlIgnoreCase(token, class_name)) {
+                return true;
+            }
+        } else if (std.mem.eql(u8, token, class_name)) {
+            return true;
+        }
     }
     return false;
 }
 
-// The class attribute tokens are separated by ASCII whitespace (which,
-// unlike std.ascii.isWhitespace, does not include vertical tab).
-fn isClassWhitespace(c: u8) bool {
-    return switch (c) {
-        '\t', '\n', 0x0C, '\r', ' ' => true,
-        else => false,
-    };
-}
-
 pub const Part = union(enum) {
     id: []const u8,
     class: []const u8,