Hi Sliderule768,
I do not own Sallies or Stillpoint Aperature II’s but have experimented extensively with room treatments with my SoundLabs, so thought my experience/thoughts might be useful/helpful.
I found that pulling the speakers out 5-6feet from the front wall behind them worked best which amounts to about a 9-10ms delay of the rear firing sound. This amount of delay falls just shy of calling attention to its self as a discrete reflection which maybe why when you have them 12 feet out polarity changes are more audible as separate reflections. More than 10ms and the precedence effect starts to break down. Three feet from the front wall would harm soundstage depth and the rear firing sound reflections are just over 5ms making them nearly indiscernible from the front firing soundwaves, hence your challenge/inability to hear polarity changes.
When my speakers were 5-6ft from the front wall, it sounded best as an untreated surface meaning no diffusion and no bass traps on the front wall behind the speakers. I focused the treatments on the rear parts of the sidewalls behind my MLP and back wall.
Are you able to take acoustical measurements of frequency response and decay times? These would serve you well if they were available and could help with finding positions where treatments were most effective.
If you are “stuck” with a 3foot distance to the front wall, you can try 6″ or deeper QRD diffusers that when positioned with their wells vertically will scatter/diffuse horizontally which aids in soundstage width.
Besides acoustic treatments and acoustic measurements, digital room correction via FIR convolution files is very effective as is multiple subwoofers, but you may already be aware of all this . . .