I got burned by this as well after upgrading to ESXi 6.7 and the latest ESXUI from the fling page.
Dies at 99% suggesting some sort of finalizing work is going bad, rather than a direct network cause. Network related logs both on the host, the switch, and the SAN don't show abnormalities. This occurs on Chrome 67 and Firefox 60. Not seeing anything in other ESXi logs to suggest what's wrong.
This is with an OVF I have imported through ESXUI successfully under ESXi 6.5 so I am reasonably sure the OVF isn't bad.
Some sort of IO timeout error from a smaller timeout value somewhere perhaps?