Ah this line with 10. Nh4, I remember that in an online blitz game, Suat Atalik played this against me. I think that I did 10...a5 and he doubled my pawns. It looks to me that White wants to play a Catalan structure where Bxc6 destroys the queenside pawn structure, but the difference is that here in Grünfeld Black has the bishop on g7 pointing at d4, with aims of ...c5 or ...e5, and White generally cannot do d5 because it might be too weak. But I could look at this and see what can do for White.
I agree also, that playing Fianchetto against that ...c6/...d5 setup is very difficult to claim an edge. I know from experience because I played this with Black as well. But Grünfeld players generally want a dynamic position like the ...Nxd5 positions instead of a semi-closed slow symmetrical game. Even if you get barely a minimum of edge of Black equalises (latter is much more likely), opponent has to play the position, which I suppose that is a bit unpersonlike for the Grünfeld player playing this Slaw-like structure. Unless of course you play both the Grünfeld and Slaw like I do
This is one of the few lines where my White and Black repertoires clash, and Black seems to have better results here. But this line is very fresh, White has been trying looks like every legal move after 9...Re8. So who knows
I wonder what Pert recommends, if it is 10. Nh4. I personally think that 10. Re1 a5 is reaching the end of a try for an advantage for White, unless someone finds something weird and unexpected. I see too that Sandro Mareco beat David Navara with 10. Nh4, and Mareco is known for his deep opening preparation. The race for ideas continues