I use a F/L die, adjusted upward to only resize about the end 1/3 of the neck length. Actually, even less; I test the seated bullet to ensure it can't be twisted in the neck using forefinger and thumb only (this is for hand fed single loading only).
Once the bolt closure resistance becomes significant, I will run the die down to bump the shoulder to nearly remove the resistance, then reset it to partial resizing length. Another approach would be to use, say, a .308 resizing die to bump-size a .260 case, leaving the case neck untouched.
I believe this practice has several advantages.
First, it reduces the length of the neck that gets work hardened, and confines it to the end, where some of it will get trimmed off in time.
Next, the rearward, unsized portion of the neck centers the case neck in the chamber better. Also, since the bottom end of the sidewall does not get resized as much, it also centers in the chamber better. This also works the casewall bottom less, possibly reducing some of the tendency to develop head separations.
Finally, the twist test ensures that there is (just) enough neck tension to prevent the bullet creeping during hand chambering. Thus ensures minimal acceptable neck tension, which I believe assists better accuracy. Because this test gets performed each reloading, it takes into account any neck brass work hardening.
This is not some hokey trick I cooked up on my own. It is an old BR handloadng technique from earlier days, when neck-only dies were less common. It was taught to me by my Elder Brother, who did BR shooting in the 1960's and 1970's.
Greg