Hello,
I had same problem and luckily I was able to solve the problem without using any external debugger.
1) The problem
I have recently purchased DEVKIT_MPC5748 and within 2days I got reset escalation that you have shown. I was playing with Watchdog timer (SWT) which ended up resetting the MCU every time it boots up.

RGM_DRET sets the 4bit threshold for ‘destructive’ reset escalation to keeping the chip in the reset state until the next power-on reset triggers a new reset sequence. SWT are one of the sources for destructive reset and my SWT blunder has caused MCU to reset 16 times (F is the defualt value) after which it holded reset line to LOW . So, one has to recycle the power again to bring back the MCU.
If you have accidentally written to HSM area then device recovery using solution I am going to mention may not happen (didn't happen to me and I don't want to try it). See this thread for more information:
https://community.nxp.com/thread/447721?commentID=952150#comment-952150
2) The Solution
When you power up the devkit, the MCU immediately enters into destructive reset escalation due to some reset sources. Mine was SWT from module, which caused MCU to reset 0xF times after which MCU as hanging in PHASE0 reset sequence. This all happens within a short period of time and when openSDA debugger tries to communicate with MCU it will see a MCU in destructive reset escalation state.
Here is the simple solution,
1) hold the RESET switch before you plug your devkit to PC (you should keep holding it)
2) upload the working elf (image that is working perfectly fine)
3) You will start seeing normal upload process which you missed
4) Once the flash image has been uploaded, you will however get same reset escalation message. But don't worry, now you can release your Reset line.
5) Now debugger will communicate with your recovered MCU.
by holding the reset line before power on you are ensuring that mcu will never execute your problematic flash image. Then mcu will not be entering destructive reset escalation.
3) For future
If you don't want destructive reset escalation to happen. You can write DRET field of RGM_DRET to zero immediately after boot up. This will disable the 'destructive reset escalation' feature (I have not tried it, there may be more configurations).