The simplest answer to your question is to use Berger's twist rate calculator and see where various twist rates will put you in the approximate velocity range you expect out of the shorter barrel. If this is a "practice" rifle, it may not be necessary to use a faster twist barrel to wring every last possible shred of BC out of the bullet (i.e. an Sg of 1.5 or greater). If the bullet exits the bore with an Sg of approximately 1.1 or better, it probably won't keyhole until it slows sufficiently downrange, if even then. Anywhere in the Sg = 1.3 to 1.4 range will provide plenty of stability, even if you aren't netting the full intrinsic BC of the bullet.
There is a big difference between gyroscopic stability and obtaining the full inherent BC of the bullet. The first of these obviously requires an Sg of greater than 1.0 to maintain stability even at short range. However, rotational velocity of bullets slows far less than linear velocity. As such, bullet gyroscopic stability actually increases as the distance increases, right up until the point the bullet slows to trans-sonic and below, or is overcome by dynamic instability.
Spinning a bullet fast enough to obtain the full intrinsic BC is really only necessary if you're actually competing with the rifle and need to get every bit of performance possible. In the case of heavy .224 bullets, the risk of jacket failure increases significantly as the twist rate increases. Relative to some other calibers, a 7.0-twist is actually quite fast and can put significant strain on the jacket. The .224 bullets in the 90-95 gr range are especially susceptible to this issue because they generally have very long bearing surfaces relative to lighter, shorter bullets. Running the 90 VLDs at 2820 to 2850 fps, there seems to be a very good correlation to observed jacket failures and barrel twist rates. As long as you stay at 7.0-twist, there is rarely an issue. Those that are using 6.8-, 6.7-, or 6.5-twist barrels at those velocities have noticed a higher rate of failure. That basically translates to a problem running the 90 VLDs over about 300K RPM. Even in F-TR midrange competition, where you really need to get every bit of performance out of a .223 to run with .308s shooting much higher BC pills, a single jacket failure is not worth the teeny little bit of BC you get in return. One jacket failure and you're out of the match...period.
For a practice rifle, I'd be looking at what twist rate is predicted to give you an Sg of around 1.3 to 1.4 at your expected velocities, at the location(s) you shoot most often. That should be plenty for practice purposes. For example, a 90 VLD at 2600 fps from a 7.0-twist barrel is predicted to have an Sg of 1.38 (65 degrees, 500 ft elevation), which is plenty. The only thing you would ever notice would be slightly greater wind deflection at distances past 300-500 yd, which shouldn't be a deal-breaker.
