Not really, no. Even if it were 100% purely diegetically presented--e.g. the guard puts up a perfect front and the player tries every plausible method and consistently fails no matter how high they roll--I would still have a problem with it, because it reads plainly as a violation of clear and established fiction. This is a priesthood all about defending faux-Egypt from external threats. The party has good evidence that an extreme external threat is imminent, and a person whose established backstory warrants some minimal degree of respect, even if that respect earns very, very little.
If I may, an alternative I would accept would be something like (summarizing, rather than presenting in "narration" as it were): The party speaks to a few different guards, several of whom are very standoffish or refuse to interact with the party in any meaningful way. Eventually, they find one who is very skeptical, but not so totally against any discussion--perhaps a pragmatic "if we let them think they had their say, they'll go away" kind of thing. So this guard arranges a purely informal, "this never happened" type meeting with a low-level functionary in the priesthood. That priest listens to what the party has to say, and then (without rolling) basically just stonewalls them, or maybe asks for conclusive proof, or something of the sort. In other words, the PCs were given a legitimate shot, they even got to kinda-sorta-ish speak with the priesthood proper, but they were still shut down. That would be good reason to think "hey, there's something really weird going on, these guys are supposed to be ALL ABOUT slapping down the enemies of Kemet with extreme prejudice, why are they suddenly demanding incontrovertible proof before they'll even look into it??" That would then, quite naturally, lead to an adventure to find out what the hells is going on with the Sutekh-Garyx priesthood--rather than a lead simply shut down with no explanation beyond "you have to trust me", it would actually show that even the priesthood itself is, or at least seems to be, behaving VERY weirdly and should probably be investigated.
Presume it makes no difference. The GM is narrating that their loyalty is absolute, so the characters know there's no point. How precisely they might have this knowledge is not, to my mind, particularly relevant.
I don't see how "not played out but fully carried out" doesn't still qualify as being vetoed. That's like saying a character can present all of their descriptions of how they attack an enemy in a combat, use up their resources etc., and then the GM says, "All of that fails. Next?" That's still a veto in my book.
The point was precisely that it was something potentially (not guaranteed, but potentially) coming from an acceptable motive, but was (again, by construction) something that LOOKS really, really hinky. Like, in many other contexts, that exact action would be seen as pretty blatant railroading. Here, the player is giving the GM the benefit of the doubt, but (more or less) saying "it simply isn't acceptable to do this and then tell me 'you just have to trust me', that's ignoring my legitimate concerns and asserting that, simply because you're the GM, I have to go along with whatever you say no matter what."
I disagree--quite strongly, actually, because there the player is making a declaration while in ignorance, while here, the player was making a declaration specifically on the back of both GM-approved character backstory, and GM-established context and lore. That's a pretty significant gap, since your argument here is specifically deriving its force from the fact that the player is making a declaration of which the character is ignorant. Ranakht is not ignorant of the doctrines of the various priesthoods. He is not ignorant of his (extremely minor but nonzero) position in society. He is not ignorant of the desires and interests of the Hyksos to retain their authority and their influence in society. His proposal is confident because it is built on the back of substantial knowledge, not on a lack of information about alternatives.
Well, the point wasn't "GM knows this can't work and tells the player so". The point was that the GM knowing this can't work looks a hell of a lot like railroading. It's a plausible plan, supported by good evidence, and by other things the GM has explicitly approved and perhaps even allowed the use of in the past. A sudden reversal on something like that, without any explanation beyond "you just have to trust me", is a party foul.