Allow for ignoring encoding or setting default encoding

I have a file that reports in its prolog definition that it's UTF-16, when it is indeed UTF-8. I've had to copy the parser class and do some ugly hacks to override the default behaviour (which is to fail when a weird encoding is encountered):

` override def verifyAndSetXmlEncoding(): Unit = {
val enc = CharsetNames.normalize(_textBuilder.contentsAsString)
_config.setXmlEncoding(enc)
/* 09-Feb-2011, tatu: For now, we will only accept UTF-8 and ASCII; could
* expand in future (Latin-1 should be doable)
*/
if ((CharsetNames.CS_UTF8 != enc) && (CharsetNames.CS_US_ASCII != enc)) {
_config.setXmlEncoding("UTF-8")
}
}`

It should be possible to override this behaviour and just set the encoding manually. In this case I have no control over the file, so changing the file is not an option.

Otherwise, great library :)

~Karl