After a little investigation, I believe the 2nd curve have some slightly uneven area that is not perfectly in planar even though the curve pass the Planar test.
By testing 100 points on the 2nd curve to the plane you projected to, find out that some distance is not 0.
Thank you for actually trying. I appreciate it. I had tried this as well or a slightly different way where I projected both to the same plane as opposed to intersecting. I agree with your insight that there is an uneven area somewhere, which shows up even if both are projected to the same plane and then that plane is used as boolean plane.
GH bulls**t hehe given the same discrepancy DOES allow boundary surface to work, which works with planar curves only. Oh well…brute-force workarounds…