what happens to PCM when it is processed by DTS or Dolby?

Decoded is not the right work.  "PCM" does not decode to a surround format.  It's more like "translated".  It's artificially creating a surround type music experience from 2-channel audio.  It's like "PCM" is like a picture of a two trees in front of you.  Then you as an artist need to imagine what it looks like in the middle of a forest.  That's what these DTS Neural and Dolby Music surround modes try to do.

Netflix can be PCM 2-channel or 5.1 Dolby Digital depending on the show.

HBO Max and Disney are not Dolby Atmos.  Most likely, they are all just Dolby Digital.

Anything coming through cable or satellite is usually Dolby Digital.

Sounds like either a configuration in your audio settings for Netflix on Apple TV or it is a limitation of the app.

PCM cannot be processed by DTS or Dolby Digital.  Both DTS and Dolby Digital are proprietary multi-channel encoding formats.  They are like a different language.

PCM is also NOT a compressed series of ones and zeroes.  It is uncompressed raw 2-channel data of the music in the original sampling format (such as 16 bit / 44.1 khz or 24 bit / 96 khz).

However, I think you are talking about surround decoding processes for music.  The "DTS Neural" and "Dolby Pro Logic" are processes that try to translate 2-channel audio into a multi-channel output.  I don't think anyone here can further explain in detail what happens because it's a lot of rocket science math.  You should probably google more on that specific format, such as "DTS Neural".