Let's just say the Lishen cells will reliably deliver at 50C, that's 900A, if we have 18AH capacity you will have a little over 1 minute reserve. My guy in China says these high C rating Lishen cells are the popular ones now for car audio applications, but I have not seen the discharge graphs for these. I would expect they start dropping voltage and current potential noticeably below 40% charge so real world I don't know what you would really expect by way of how long you could actually beat on things with so little reserve.
I would avoid the headway in general. If you must use old fashioned LiFePO4 see if you can't find some c-max cells instead. Those have a more proven track record for our applications.
I will say that LTO has far better performance in extreme cold. Published ratings are good and real world performance up here in Maine has been good.
I would suggest against mix matching batteries. If you can't afford to do the electrical system right you shouldn't be considering the type of build that will demand 800-1000A current. Keep in mind, perfect world, the difference between 5000W and 10,000 is only 3dB which is only 2 clicks on the volume knob and your expense all around will dramatically increase for a maybe gain.
As to how long LTO really lasts in our applciations it is far too early to say. There were some in use in Russia perhaps 4 years ago, my brother and I built our banks about 3 years ago now and AFAIK we were the fist ones in the USA using these for car audio so there just isn't any real world data for us yet to confirm or deny the 10-20-30 year claims. That said, LTO isn't exactly new and those cycle life numbers are likely pulled at least partially from EVs and backup power supply banks so they're probably at least close to accurate.