HTML API: Add method to split text nodes by null or empty prefixes.

[dmsnell]

There are places in the HTML Processor that need to parse differently according to whether text content is a sequence of NULL bytes or whitespace characters after decoding. It's awkward and inefficient to do this within the HTML Processor, however, as it requires eagerly decoding text nodes.

The Tag Processor could expose a method to efficiently split apart a text node when needed, and then classify it, to aid in the parsing. This method could further be used to identify inter-element whitespace, which is usually ignored when rendering HTML.

Trac ticket: Core-61974

HTML parsing rules at times differentiate character tokens that are all null bytes, all whitespace, or other content. This patch introduces a new function which may be used to classify text node sub-regions and lead to more efficient application of these parsing rules.

Further, when classified in this way, application code may skip some rules and decoding entirely, improving performance.

## Example script

<?php

require_once __DIR__ . '/src/wp-load.php';

$p = new class( "\x00\n\r&#x000000020;\n\x00\x00 \tStuff\x00\f<p>&#x13; &#9;\ntext<p>a \x00b<p>\x00\x00" ) extends WP_HTML_Tag_Processor {
        public function get_token_length() {
                $this->set_bookmark('here');
                return $this->bookmarks['here']->length;
        }
};

while ( $p->next_token() ) {
        if ( '#text' !== $p->get_token_name() ) {
                echo "\e[3;90mSkipping \e[0;2;32m{$p->get_token_name()}\e[m\n";
                continue;
        }

        $did_split = $p->subdivide_text_appropriately();
        $text = $p->get_modifiable_text();
        $text = str_replace( [ "\x00", "\t", "\f", "\r", "\n" ], [ '␤', '␉', '␌', '␍', '␤' ], $text );
        $after = $did_split ? " \e[3mafter splitting" : '';
        $length = $p->get_token_length();
        echo "\e[90mFound (\e[33m{$length}\e[90m) '\e[34m{$text}\e[90m'{$after}\e[m\n";
}

## html5lib tests

-Tests: 2433, Assertions: 4116, Skipped: 421.
+Tests: 2433, Assertions: 4112, Skipped: 425.

sorry for the late ticket creation, but I thought it was best to separate this as a feature enhancement. it's mostly an internal method, but since it has potential use to application code it can retain its own ticket.

[dmsnell]

In 58970:

HTML API: Allow subdividing text nodes by meaningful prefixes.

HTML parsing rules at times differentiate character tokens that are all null bytes, all whitespace, or other content. This patch introduces a new function which may be used to classify text node sub-regions and lead to more efficient application of these parsing rules.

Further, when classified in this way, application code may skip some rules and decoding entirely, improving performance. For example, this can be used to ease the implementation of skipping inter-element whitespace, which is usually not rendered.

Developed in ​https://github.com/WordPress/wordpress-develop/pull/7236
Discussed in https://core.trac.wordpress.org/ticket/61974

Props dmsnell, jonsurrell.
Fixes #61974.

Merged in [58970]
​https://github.com/wordpress/wordpress-develop/commit/95eb879c47c52413f33c3a62f3006262cc2b0062