You indicated: "BTW, a feature of such a structure might include a judgement about the degree of confidence of relationships - similar to the radio buttons for sources:"
I agree 100% and so does GEDCOM.
As a part of all Source_Citations a QUAY or Quality indicator is available, which is defined as:
"The QUAY tag's value conveys the submitter's quantitative evaluation of the credibility of a piece of information, based upon its supporting evidence. Some systems use this feature to rank multiple conflicting opinions for display of most likely information first. It is not intended to eliminate the receiver's need to evaluate the evidence for themselves."
And while it does not use your words, remember, it is the source that gives the evidence that a fact (or in this case an association) is correct and you (and the reader) must determine if the source/evidence is really good, kinda good, questionable, or rotten.
I by the way disagree that the Quality flag should say "proven by DNA" because the Source should be the DNA Lab Results and then the quality of the source should be "is the source very good, kinda good, questionable, or rotten?" In most cases DNA would be a "3 = Direct and primary evidence used, or by dominance of the evidence".
But then again, a DNA sample and result from 10-15 years ago would be less reliable than a sample and result from today. And a botch collection or an unreliable lab may send your DNA proof into the relm of "Rotten".
So while the GEDCOM could use a little updating in some areas, I'm not convinced that this in one of those areas.