contentType effects for test writers (build updated for HTML L2)

I want to recap the implications of the contentType work on the tests.

All test cases should NOT specify an extension on the <load> element, but allow the extension to be determined by the contentType.  All existing cases in the CVS have been updated to remove the .xml extension from the <load> element.

Some tests cases that target both HTML and XML should use ignoreCase="auto" to allow the case-sensitivity to vary depending on the contentType.  Typically, these would be cases that check the nodeName or tagName of selected node.  The expected value should be lower-case.  For example:

<assertEquals actual="docElemNodeName" expected='"html"' ignoreCase="auto" id="nodeName"/>

A very small handful of cases will need to have conditional sections based on the contentType being processed.  The most common are tests that check the document element name which is typically "svg" for "image/xml+svg" instead of "staff".

For each recommendation, there should be contentType specific alltests suites.

If you find that you need to write a lot of stuff in the test to adapt to contentTypes, then I haven't done my job well.

p.s. I've updated build.xml to point to the new HTML working draft.  You will need to manually remove any existing DOM2-HTML.zip file from lib/specs since the new "offline" behavior will not update a previously downloaded specification zip file.