Use SWAR to seek for non-ASCII UTF-8 in DOM parsing (#16350)

GitHub FYP test case:
```
Benchmark 1: ./sapi/cli/php test.php
  Time (mean ± σ):     502.8 ms ±   6.2 ms    [User: 498.3 ms, System: 3.2 ms]
  Range (min … max):   495.2 ms … 509.8 ms    10 runs

Benchmark 2: ./sapi/cli/php_old test.php
  Time (mean ± σ):     518.4 ms ±   4.3 ms    [User: 513.9 ms, System: 3.2 ms]
  Range (min … max):   511.5 ms … 525.5 ms    10 runs

Summary
  ./sapi/cli/php test.php ran
    1.03 ± 0.02 times faster than ./sapi/cli/php_old test.php
```

Wikipedia English homepage test case:
```
Benchmark 1: ./sapi/cli/php test.php
  Time (mean ± σ):     301.1 ms ±   4.2 ms    [User: 295.5 ms, System: 4.8 ms]
  Range (min … max):   296.3 ms … 308.8 ms    10 runs

Benchmark 2: ./sapi/cli/php_old test.php
  Time (mean ± σ):     308.2 ms ±   1.7 ms    [User: 304.6 ms, System: 2.9 ms]
  Range (min … max):   306.9 ms … 312.8 ms    10 runs

Summary
  ./sapi/cli/php test.php ran
    1.02 ± 0.02 times faster than ./sapi/cli/php_old test.php
```

ext/dom/html_document.c

@@ -30,6 +30,7 @@
 #include <Zend/zend_smart_string.h>
 #include <lexbor/html/encoding.h>
 #include <lexbor/encoding/encoding.h>
+#include <lexbor/core/swar.h>
 
 /* Implementation defined, but as HTML5 defaults in all other cases to UTF-8, we'll do the same. */
 #define DOM_FALLBACK_ENCODING_ID LXB_ENCODING_UTF_8
@@ -517,6 +518,30 @@ static bool dom_process_parse_chunk(
 	return true;
 }
 
+/* This seeks, using SWAR techniques, to the first non-ASCII byte in a UTF-8 input.
+ * Returns true if the entire input was consumed without encountering non-ASCII, false otherwise. */
+static zend_always_inline bool dom_seek_utf8_non_ascii(const lxb_char_t **data, const lxb_char_t *end)
+{
+	while (*data + sizeof(size_t) <= end) {
+		size_t bytes;
+		memcpy(&bytes, *data, sizeof(bytes));
+		/* If the top bit is set, it's not ASCII. */
+		if ((bytes & LEXBOR_SWAR_REPEAT(0x80)) != 0) {
+			return false;
+		}
+		*data += sizeof(size_t);
+	}
+
+	while (*data < end) {
+		if (**data & 0x80) {
+			return false;
+		}
+		(*data)++;
+	}
+
+	return true;
+}
+
 static bool dom_decode_encode_fast_path(
 	lexbor_libxml2_bridge_parse_context *ctx,
 	lxb_html_document_t *document,
@@ -534,13 +559,13 @@ static bool dom_decode_encode_fast_path(
 	const lxb_char_t *last_output = buf_ref;
 	while (buf_ref != buf_end) {
 		/* Fast path converts non-validated UTF-8 -> validated UTF-8 */
-		if (decoding_encoding_ctx->decode.u.utf_8.need == 0 && *buf_ref < 0x80) {
+		if (decoding_encoding_ctx->decode.u.utf_8.need == 0) {
 			/* Fast path within the fast path: try to skip non-mb bytes in bulk if we are not in a state where we
-			 * need more UTF-8 bytes to complete a sequence.
+			 * need more UTF-8 bytes to complete a sequence. */
-			 * It might be tempting to use SIMD here, but it turns out that this is less efficient because
+			if (dom_seek_utf8_non_ascii(&buf_ref, buf_end)) {
-			 * we need to process the same byte multiple times sometimes when mixing ASCII with multibyte. */
+				ZEND_ASSERT(buf_ref == buf_end);
-			buf_ref++;
+				break;
-			continue;
+			}
 		}
 		const lxb_char_t *buf_ref_backup = buf_ref;
 		lxb_codepoint_t codepoint = lxb_encoding_decode_utf_8_single(&decoding_encoding_ctx->decode, &buf_ref, buf_end);