Those Highpoints are "rolled gold" - whatever you do don't blame or question this device (SSD7101A). It's a quality PCB with a PLX-brand chip and it works with everything by design. There's a known issue with the ASM2824 but not with PLX switches. Before lamenting the failure of this card I would point to many more obvious suspects:
(1) APFS is a much more likely suspect. Everything works better on HFS+ (including the ability to build a bootable RAID across 4 NVMe disks WITHOUT 3rd party kexts)
(2) Your NVME disk brand and firmware (you are only guaranteed perfect compatibility and max transfer speeds with a Samsung 970 PRO, 970 EVO, SSUBX or SSUAX - beyond that - the chances of your NVMe SSD being the culprit rise exponentially)
(3) When everything "broke" (spinning wheel), your previous bootargs became redundant/useless. You'd have to reset PRAM.. and then fix everything from a SINGLE basic boot drive (like you did with the MOJAVE HDD).. and then reintroduce your HPT SSD 7101 (only after one cold shutdown and one clean restart with the basic boot drive)
(7) the Highpoint card is defective (it's not and I feel sorry for the non-english speaking electrical engineering wizards at Highpoint that can't possibly articulate the sheer number of dependency failures that should be examined before they are obliged to make good on a refund request).
(1) APFS is a much more likely suspect. Everything works better on HFS+ (including the ability to build a bootable RAID across 4 NVMe disks WITHOUT 3rd party kexts)
(2) Your NVME disk brand and firmware (you are only guaranteed perfect compatibility and max transfer speeds with a Samsung 970 PRO, 970 EVO, SSUBX or SSUAX - beyond that - the chances of your NVMe SSD being the culprit rise exponentially)
(3) When everything "broke" (spinning wheel), your previous bootargs became redundant/useless. You'd have to reset PRAM.. and then fix everything from a SINGLE basic boot drive (like you did with the MOJAVE HDD).. and then reintroduce your HPT SSD 7101 (only after one cold shutdown and one clean restart with the basic boot drive)
(7) the Highpoint card is defective (it's not and I feel sorry for the non-english speaking electrical engineering wizards at Highpoint that can't possibly articulate the sheer number of dependency failures that should be examined before they are obliged to make good on a refund request).