Eval bug in the Von Hennig Gambit, and maybe in simialair situations.

In the Von Hennig Gambit after 6 ... Bg4, the correct move according to chess.com engine, and Lichess engine 7 Bxf7+!! (It used to be classified as a brilliant move). Looking at the chess.com engine in the eval it is saying that 7 Bxf7+ is +1 however, the game review analysis tab is saying 7 Ne5 is the "Best Move" even though it's the second line it gives, and is only +0.9. IDK, seems goofy, maybe the moves are so close in eval, the engine has a hard time picking which one? Although previously it consistantly said 7 Bxf7+ was the best move. Don't get me wrong I love the "Oh no my Queen" line as much as the next Von Hennig player, but it's never to my knowledge been considered the best line in the Von Hennig

(game and screen shots for reference) https://www.chess.com/analysis/game/live/79500053767?tab=analysis

The game review engine and the engine giving the lines are different, with the game review one being stronger

Maybe, like I said though I don't think 7 Ne5 was ever considered the "better move" between that and 7 Bxf7+ I guess looking at after the move it gives Ne5 +1.53 and Bxf7+ +1.28, interesting, I suppose it wouldn't be the first time a computer changes theory, Bxf7+ has been considered "Mainline" for the Von Hennig, given this position for as long as I've known about it, interesting.

I think you mean lines one being stronger

Unless you are talking about max-depth game review (which a non-diamond member might not have)

True

Just saying they don't have the same strenght/depth, so the eval might be different for different lines/moves

Perhaps I am only platnium, but that would be a huge added value to Diamond Members, (Gold doesn't even get more than one game review anymore). That a Diamond level member would see 7 Bxf7+ as the "Best Move" or as I mentioned it used to be considered a "Brilliant Move", and Platinium members would see it as only a "Good Move", that is a huge difference in terms of studying for a Diamond level member and a Platinium level member.

Mainline theory and engine evaluations have nothing to do with each other. Evaluations can be influenced by depth and by the engine being used. The classifications are also influenced some by player ratings, some sacrifices, won't be brilliant for higher ratings.

Yeah i know like "The machine doesn't care about your theory", but like i said previously what would be considered mainline (7 Bxf7+), was evaled as better by the engine(whether it says it's brilliant is not really relvant, just an interesting note i threw in) it's just, that now the computer doesn't think 7 Bxf7+ is the better move, heck i just checked it on other multiple online engines Lichess, Next Chess Move (which uses stockfish 15.1, or you can select the nightly build),and a few others I could find, i'm assuming most of them likely using some Stockfish derivative. Like i'm pretty sure every other engine will say 7 Bxf7+ is the best move. Unless chess.com is using a learning engine, that goess based off of the players games, which could make sense, I have more wins with 7 Ne5, mainly becuase i play it more often, so I have 27 wins and 4 losses with that (87.333...% winrate) move, and only 5 wins and 1 loss (83.333...%) with 7 Bxf7+, but as I said i've played "Oh no my queen" a lot more (clearly lol) than 7 Bxf7+