If the char you want to cast time warp on goes after Fane (which can be easily arranged as apart from having 1 high ini char you just care about ini for turn order) you're only giving turn to 1 enemy, which is hardly a problem. You can only "win in one turn" with timewarping yourself if you're lonewolf or lategame anyway, in full party your Fane won't be strong enough to do that with just 1.5 or 1.66(if glasscannon) turns he has left for a while. And lategame if your main damage dealer mage goes right after Fane a good combo would be timewarp + haste on the mage with Fane and then apotheosis + nukes with mage. This indeeed allows you to "win in one turn" quite often and won't be possible with Fane himself unless he has something to source vamp, furher reducing his remaining AP. Otherwise you're just trading 1 SP for 2-4(glasscannon) extra AP on Fane, which isn't bad but hardly amazing.
You're deliberately shorting yourself and you've already lost in terms of exchange because you're giving the opposing team a turn versus having any chance of denying the enemy team a turn. So, your AP comparison doesn't really work out and the risk/reward makes virtually no sense under many circumstances.
You can win in one turn, or reach your objectives, in one turn by the time you reach Reaper's Coast. Most glass cannon builds will be able to CC the most dangerous target or take out several opponents with that Time Warp. You're literally saying that spending 2AP to deny, at minimum, one enemy and up to the entire encounter's worth is not a good trade.
Also, you Apotheosis first to reduce the cost of all actives to 0 and then use Time Warp. But in Reaper's Coast you start off every fight with one source if you put some thought or effort into it. So, why not? When you're in Act 2 having 13AP or more means that any targets caught by Fane will not have a chance to act that turn. The extra AP is pretty critical to re-positioning targets with teleport, netherswap, etc. You do plenty of damage and enough to CC almost any target by 8 (enemies relative to you).
Having Fane not go first would require your game plan to account for a lot more, have more points of failure, and would overall lose to the former where as having Fane has no draw back in the current meta. Having Fane not go first also means your entire party is not using Fane to his maximum potential either in which case we don't have much to discuss because we're looking at different parameters.
You're seriously arguing against the meta right now and saying that it is more efficient to leave enemies up and running to do their thing when you could have easily taken them down. God forbid your primary damage dealer gets hailstormed to death because you allowed them to take a turn. Or, if you were in a pvp situation and decided to let the enemy take a turn and they vaporize you for it.
Wait, wait.. since when is it ever more optimal to let someone take a turn?
