Extract <h1>…</h1> title from rendered HTML and add it as <title>…</title> tag, too

The returned HTML documents are full documents, but don't have a `<title>…</title>` tag with most formats.

It would be nice if the HTML coming from the formatting engines could be checked for (the first) `<h1>…</h1>` tag and their contents is then added as `<title>…</title>` tag, too.
