I suspect one of the points you are missing is that Germany certainly did call Einstein a traitor, and certainly had laid the groundwork for executing him specifically for treason and not just as part of the final solution if they had captured him after a successful conclusion to their war.. Another one is that the United States has a very limited definition of treason, which is actually spelled out in the constitution.
Article 3 - The Judicial Branch
Section 3 - Treason
Treason against the United States, shall consist only in levying War against them, or in adhering to their Enemies, giving them Aid and Comfort. No Person shall be convicted of Treason unless on the Testimony of two Witnesses to the same overt Act, or on Confession in open Court.
The Congress shall have power to declare the Punishment of Treason, but no Attainder of Treason shall work Corruption of Blood, or Forfeiture except during the Life of the Person attainted.
By that definition, neither Einstein nor Snowden would even possibly count as a traitor. The arguement you're making comes off as everyone should use some other definition of traitor that is broader than in the US Constitution, but somehow doesn't allow the sort of abuses a nation such as Nazi Germany would commit with such a legal basis, so that Snowden might count while Einstein, Fermi, et. al. couldn't possibly.
The point is, we have two entities who appear to fall in the same domain (non-treasonous things). Someone created an analogy that correctly asserts these two entities do in fact belong in the same domain. Then someone else declared, by apparent fiat, that the analogy was irrelevant. Not flawed, not violating some principle of logic, but simply irrelevant. Every analogy is imperfect. All are flawed to some extent, and matter only because they can still be useful to get to a correct conclusion despite the flaws. This analogy may have more flaws than a great many, (In fact, I think analogies of this sort seldom lead to the correct conclusion, and generally shed more heat than light on their subjects) but still, in this case, it has somehow led to the correct conclusion, therefore it simply cannot be irrelevant. Declaring it irrelevant is thus not a counterargument, but an attempt to suppress speech. You probably don't want to endorse the AC doing that, instead of rationally addressing the flaws specifially and not just dismissing the whole.