Hi,
This depends on what you mean by panorama, please go into a little more detail on how you are taking them.
To get a good 3d model, it is essencial that you have photos taken *from* many difernt locations around/in the scene, in effect the panorama would only add a single extra data point here, so it may not be the most efficient way to do it.
It's also essencial that the photos have not been "messed around with" in terms of perspective and geometry etc, and processed panoramas are generally heavily modified in this way, making them problematic.
A workflow that could work, depending on how you do things is to use the source (unedited) images for a panaorama and use them as some of the source images for Metashape. They would need to be added with many other "properly" taken images. The only situation where this would be useful is if you were taking a panorama as well as doing photogrammery on a scene, and it would only add a little value. You would also have ot be careful to use similar camera settings.
I personally do panorams in 2 ways, I take quick ones on my phone or using the panorama mode on one of my older DSLRs, these images would be almost imposible to use here as you never get access to the unedited images. I also take "proper" panoramas using normal images form a DSLR, and spend a while processing them in Hugin to render the result.
Thanks,
JRP