I was able to reproduce the same BIOS issue when trying to deploy an SNO node on a DELL R740 machine. Cloning this BZ as it looks like we need the same Ironic Python Agent workaround implemented in the assisted-installer ISO. Derek Higgins proposed a patch at https://github.com/openshift/assisted-installer/pull/313/
I have also run into this issue with both the 4.8 and 4.7 baremetal installers on Dell M630 nodes in our lab. My work-around at the moment is to use the system settings (BIOS) menu to disable the RHEL and ironic1 UEFI boot entries. Install is able to proceed at that point. Once the system is reinstalled, I am able to use efibootmgr -B -b <4 digit hex> to delete those same extraneous entries. I observed that while these entries are present, booting from a virtual CDROM fails. (Related to Derek Higgins' comment 61 from the original BZ https://bugzilla.redhat.com/show_bug.cgi?id=1966129#c61)
https://github.com/openshift/assisted-installer/pull/313
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory (Moderate: OpenShift Container Platform 4.9.0 bug fix and security update), and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHSA-2021:3759