AAPT2: XML configuration file parser.

The XML file is parsed with the aapt::xml::XmlDom parser and actions are
used for the elements we are interested in. This means that there are
cases where a user could add aditional tags/elements to the document
that are technically invalid (acording to the schema) but are ignored by
the parser. This allows us to be more lenient which probably isn't a bad
thing.

Documents have the namespace stripped before processing as the
XmlActionExecutor ignores any elements with a namespace. The namespace
is validated before being removed.

The test cases are all based off the previous example XML file.

Test: Unit tests

Change-Id: I86d7e0dc6347ace3eaa60c1842d59f3cd0d4f749