While cone error can be corrected with a good mount model to improve GoTo's. It is desirable to eliminate as much mechanical misalignment as possible.
I would like to see a tool added to the alignment routine similar to the Polar alignment assistant, but for cone error.
PC Software ConeSharp implements a strategy that could be adapted and improved upon with plate solving.
The steps to do it would involve first a precise polar alignment with the Polar Alignment tool. Then to measure cone error slew to a star with low DEC west of the meridian, solve, and record pointing error. Then slew east of meridian, solve and record that error. Lastly calculate a correction vector based on recorded pointing errors.
Is there any interest for such a function?

Cone error is fairly easy to measure with any of the polar alignment tools that work in the polar region. Once you identify the centre of rotation; any cone error is the component of the offset of the centre of rotation fom the centre of the sensor that is orthogonal to the declination direction. It is unaffacted by polar alignment. The Static PA module I just added to PHD2 has the functionality to display cone erro but is disabled. Something similar could possibly be added to Ekos polar alignment

That is very interesting. However I generally use my guide scope to perform the PAA because it is much faster at transferring images than my imaging DSLR. While that is ok because it relative to the mount rotation, it's orthogonality can be way off from the OTA; thus giving a different cone error. Therefore it's own routine, strictly imaged through the OTA is necessary.