Yes, crosstalk exists in all devices in the audio chain. It's usually irrelevant in line level parts like DACs and most
preamps. Amir does not measure it for DACs, but does so for preamps. You won't hear crosstalk at -100 dB. Even at -80, I would be sceptical, especially if it's outside of the
most sensitive region of hearing.
As explained, it affects all devices but relevant levels of crosstalk can usually only be found in power amplifiers. I would assume that the reason is the much higher voltage found in amps, which causes the electromagnetic field in the PCB tracks and cables to extent much further.
Why would phase changes "smear transients" or "affect spatial cues"? If you select a minimum phase filter on a DAC, it sounds the same as the corresponding linear phase one. However, the phase/frequency relationships of both filters look quite different.
As far as I am aware,
phase isn't usually of concern for audio reproduction. Heck, for the majority of recordings, it's not even clear if their phase has been flipped at some point during the mastering process. We all listen to phase-inverted music all the time and nobody can tell.
Folks also believe that there is an audible difference if they switch power cables. People are subjective and believe the most ridiculous shit. Doesn't make it true.