A lot of work...
"more of just a lifeless mannequin" -- to fool an autopsy you need a lot more than that.
Possibly not even a clone might be enough; epigenetic factors apart (which influence things as fingerprints), things like micro-traumas or calluses or muscle development require that the body has been exercised in the appropriate ways.
3D printing at the cellular level could be enough; the cellular biochemistry would probably be off, but this is not the kind of thing that's checked in an autopsy.
Note that if you use "advanced technology that's obviously available in any one age", then it is likely that whoever performs the analysis is not fooled by the dummy. Think Photoshop: a 2017-photoshopped photo (or doctored CGI video) could seem authentic to an expert from 1960, but that's because he doesn't know what CGI can do. A contemporary expert would most probably never be fooled.
So what you want is a clever enough use of not too advanced technology; or some technology that's not yet mainstream (for example, the inventor of cellular 3D printer could fake his own death if he doesn't reveal what he has invented. If he does, his customers probably would never be able to).
...could turn out to be unnecessary
Much thought should be dedicated to organizing the demise of the character. If things are set up correctly, you might need very little apart from some blood and cellular samples, and some creative scenario in which the body is beyond all doubt destroyed utterly and reliably -- except for those samples.
You could for example clone a hand, amputate that hand and replace it with the cloned one. The amputated limb would pass every conceivable test since it is authentic, and after some rehab, the "new" hand would be as good as the old one.
Applying authentic limbs to an unrecognizable and unanalyzable corpse (destroyed by hard radiations, aggressive chemicals, fire and so on) could make a strong case for the DNA's owner being very dead; detecting that the limbs were attached to a different body might not be very high in the anatomopathologist's to-do list.
Harvesting several months' of blood donations and conserving them without chemicals (if possible, given the technology) would allow stunts like simulating a blood loss from which no one could conceivably survive, not even if immediately administered transfusions.