I fully understand how they work and use them for non drone in daily life.
We have 2 simple facts here - the DJI manual states calibrate if moving over 30 miles or unused for 30 days and (ii) the drone appears to be knowing its moved a significant distance (probably via gps) and is prompting a calibration. You can deny it all you want but thats whats happening. See also the calibration after 30 days thread with someone getting that from the .dat file.
It looks like they've changed how it works and the checks it does to fit in with the new manual revisions.
(or are you also claiming DJI themselves and the engineers are unaware how their own compass works despite building it?)
The 30 day trigger appears to be real, since it gets recorded in the event log:
34.367 : 1922 [L-COMPASS][mag_cali_pt] date_from_last 30|
34.522 : 1931 [L-FMU/LED]action changed. compass out of range:need_cali(1)
The firmware certainly could trigger a request based on current location relative to last calibration location, but it does not appear to be doing that in the sense that (a) it doesn't consistently make that request after changing location and (b) there is no location check apparent in the event log.
As for the DJI statements on the issue it's not that the engineers don't know how these things work. The DJI manuals and statements themselves have been contradictory on this since the P2, and so we know that whoever writes the manuals doesn't always understand the subject.
I asked a few posts back if you would be willing to explain the nature and reasoning behind the calibrations that you mentioned as being required for other compasses you use. I'm still interested to hear that because I think it would advance the current discussion.