Replace HTMLParser.unescape() with html.unescape()

HTMLParser.unescape() was removed in Python 3.9. Use html.unescape() instead. Only fires where the parser's type resolves, and leaves the call alone where html already names something other than the module, or where the parser is an expression with side effects, such as a call, that the rewrite would drop. A call holding a comment is left alone too.