That literally can't be true.... if random is defined as "each card has an equal probability of ending up in any given position", then the only way the starting state can have any impact on the number of shuffles needed is if the starting state is already at least partially randomized. And purposely keeping lands apart from each other is the opposite of randomized.

Pileshuffling doesn't increase randomness, but it does increase entropy, which contributes towards the desired goal, which is not "each card has an equal probability of ending up in any given position" because that's outright impossible to achieve through physical shuffling, but something more like "each player has a fair chance of getting good draws or bad draws and can't predict the order of the cards". Specifically, pileshuffling makes it significantly more difficult to predict the order of the cards, so it's neither pointless nor cheating.

If you sort your deck into lands and non-lands and then pileshuffle so that the lands end up evenly distributed, that's cheating because it gives you a better than fair chance of good draws. Pileshuffling in order to evenly distribute the essentially-random selection from your deck that is the cards you drew during your previous game doesn't really make your draws better or worse on average so it's not cheating.

To illustrate the point, I might remember that I drew my Force of Will off of a Brainstorm and then played it immediately afterwards last game, resulting in those two cards being on top of one another in my graveyard, resulting in those two cards being on top of one another in my deck before I shuffled it, resulting in a greater chance that they're still on top of one another after I've shuffled it pretty thoroughly but not perfectly because it's never perfect no matter how long you shuffle it.

If I pileshuffled that deck before shuffling for real, the cards end up exactly 10 cards away from one another, and the odds of those cards still being exactly 10 cards away from one another after the actual shuffle is significantly smaller because there are more chances for other cards to get in between and there are more chances for you to completely separate those cards. You could argue that the original problem with the FoW and the Brainstorm now applies with the FoW and the card that was exactly 6 cards away and is now on top of it after the pileshuffle, but in practice, you're never going to remember what card was exactly 6 cards away from the FoW unless you cheat on purpose, so that doesn't matter.