It's been a while since I tried getting a Mac running Qubes. A lot of the time, the problems with getting Qubes running are the same problems with getting Linux running on a machine. You might want to see this article from Nov 2017:
https://www.phoronix.com/scan.php?page=article&item=mbp2016-linux-wreck&num=1
and this gist (for the comments for that article) detailing various problems and their solutions in getting your specific model working with Linux:
https://gist.github.com/roadrunner2/1289542a748d9a104e7baec6a92f9cd7
Another thing, at least with a MacBook Air 2012 5 1/2 years ago, I had to partition things in a particular way to get EFI boot to work -- rEFInd alone was not enough. Otherwise, I saw saw something similar -- installer ran, but Qubes would not boot. Details of that were posted here (not sure what happened to the original post, but it can be seen in the followup posts):
https://groups.google.com/d/msg/qubes-devel/uLDYGdKk_Dk/FsnYulsNG50J
I think I have seen Mac users in later years use that to get Qubes booting. I would start along those lines first, as digging into other things will likely be harder.
If you want to remove some of the Qubes-specific variables from this process, such as the use of virtualization features, maybe start with trying to get Fedora installed and running first.
Finally, I learned that the ACPI tables are not always 100% correct with the Macs. Specifically, I encountered a bug in a ACPI DMAR table that kept Qubes from using Vt-d. It would boot, but Vt-d was not enabled:
https://groups.google.com/forum/#!msg/qubes-devel/uLDYGdKk_Dk/aV4QncBokooJ
I eventually worked up a hack to get Xen to work around the faulty table to get Vt-d enabled, but investigating and solving something like that is well outside of the usual "having trouble getting Linux working on my laptop" types of woes.
Good luck,
Eric