Regarding the 1T45s blowing out other chips: you mention that this happened with dody placement/soldering. This could very well be caused by shorts. Consider for example a short between pin 5 (DIR) and pin 6 (VCCB/VREF_OUT) of the 1T45. There is no resistor on the DIR line to the RP2040, so if you set VREF_OUT to 5V in that case it will blow the RP2040 for sure. The resistor between the 1T45 and the outer IO connector of the Bus Pirate doesn’t matter at all in this case. Other variations of shorts with similar consequences are possible too. I think in most cases of the 1T45 killing other chips on the Bus Pirate itself, the resistor to the outer connector isn’t involved.
In my line of thinking the job of these resistors is to mostly protect the 1T45 from outer influences, things like ESD, shorting an output or the user accidently injecting a higher voltage. Another job of them is to dampen reflections and act as series termination.
The job of protecting the 1T45s could be greatly improved by adding TVS diodes.
Regarding protecting the DUT: I think this is best left to the overcurrent protection of the power supply as this is much more flexible and also includes protection for the voltage output.
I vaguely remember discussing the IO design of the Glasgow with you some time ago somewhere. But I don’t have a link to the discussion at hand.
We tested the IOs of the Glasgow with shorting them for many hours at 5V. also overvoltage from outside, but I don’t remember how hight we could go until the 1T45 blew. Another test was using the spark from a gas lighter as a crude replacement for ESD.