"People need to be aware that there is a range of models that could explain the observations,
I have to agree with that. In methodological naturalism many models will be proposed, some will gain prominence on their merits, and some won't. There is a problem that some models will become so prominent that people will stop looking for others, and Ellis' proposition serves as a good warning against that. I guess you could try to call this problem one of philosophy, but I just call it laziness.
For instance, I can construct you a spherically symmetrical universe with Earth at its center, and you cannot disprove it based on observations.
Ellis' model explained very well why we would see redshift due to gravitation. But where does Andromeda blueshifting fit into this?
Well, that's always the trick, isn't it? If you can craft the definitions so that your position is true 'by definition', then it's 'true', right?
I'm trying to explain how the philosophy of naturalism infects all of science (both methodological and metaphysical) such that it can only return a 'natural' answer and excludes a supernatural explanation 'by definition'. This is making your position 'true' by definition and excluding all other answers by definition. That is a philosophical position, however, not an empirical one.
"I guess you could try to call this problem one of philosophy, but I just call it laziness."
You still aren't seeing it. It's not which model will prevail, it's that some naturalistic model will always prevail. Always. Always. This is due to the 'a priori' commitment to the philosophy of naturalism, nothing more.
"Ellis' model explained very well why we would see redshift due to gravitation. But where does Andromeda blueshifting fit into this?"?
Here you make the assumption that redshift/blueshift is a function of recession/approach velocity that is probably not correct. Halton Arp did a lot of work noting objects with physical connections and different redshifts until they took away his telescope for heresy. (reference back to Ellis' statements) Such is science.