Description:
------------
From looking at the implementation[1], tidyNode::isHtml() is
presumably supposed to check whether the node is a start, end or
empty tag.  However, it uses members of the TidyNodeType enum[2]
as if these were bitfields, which they are not.  Therefore it
matches all node types except TidyNode_CDATA.

The documentation states[3]:

| Checks if a node is part of a HTML document

This does neither match the implementation, nor what the method
presumably is supposed to do.

[1] <https://github.com/php/php-src/blob/php-7.3.0RC3/ext/tidy/tidy.c#L1785-L1794>
[2] <http://api.html-tidy.org/tidy/tidylib_api_5.0.0/tidyenum_8h.html#a4af7ceeebf0ce18b70ed74f5db2cdf72>
[3] <http://php.net/manual/en/tidynode.ishtml.php>