I don't know but suspect quite a few... Not sure they are really "shunning" them nor what constitutes "great manufacturers", but that aside here are some points to ponder:
1. Conventional linear-phase filters offer good impulse response but by nature are symmetric in the time domain, which leads to the oft-discussed "pre-ringing" that happens as a result of making them causal. Non-casual filters, well, we haven't quite solved those yet, maybe after we get the warp drive.
2. Conventional filters already live in a lot of units and many folk do not seem to mind them. Of course, they may not know better… Manufacturers may stick with what works.
3. With oversampling DACs and better up-converting algorithms, you can move the filter’s corner up and reduce the roll-off, significantly reducing the ringing linear-phase filters add.
4. Apodizing filters add some pretty severe ringing of their own, and I wonder if the relatively large post-ringing isn’t as annoying as low-level pre-echoes but do not know.
5. A combination of apodizing and linear-phase, with or without oversampling, would seem to be the ideal solution to me. Assuming no oversampling, you’d get a bit of phase shift at the highest frequencies, a little more roll-off at say 18 - 20 kHz and up, and the advantage of the quasi-brick-wall linear phase filter by Nyquist (about 22 kHz for CDs or other 44 kS/s systems).
FWIWFM - Don