My scenario isn't exactly like yours but similar. I'm using Cisco CML for the vMX (VCP and VFP).
My default node definitions are (Ext and Int, then 4 Ge interfaces). If I added any more Ge interfaces to my vMX configuration, the FPC would either not show up, or be stuck in testing.
I noticed in my VFP messages logs, their was an "insufficient CPU" warning - even though I am using the recommended CPU/Memory, and the guest OS confirms it see's these. Instead of playing with the CPU's / Memory, I changed the VCP to lite-mode, commit , and REBOOT.
This appears to be a workaround for me - maybe you can give this a try on your end.