It's getting over complicated
We have no symptoms of a crankshaft sensor
It's not cutting out at high revs, ECU is always getting a signal for it to still start on easy start
The misfire cylinder changes when coils were swapped got to be a big hint here, yet we are ignoring it
Running with misfire is dangerous if it's a short in the coils, high voltage can and does back feed into the ECU, sounds like the original poster has got away with it for some time, not to mention the dangers of bore wash
We haven't asked basic question
What are the spark plugs like, colour, gap, when were they last changed, are any a different colour, are any damp with fuel, oil or carbon fould
Is the MAP sensor covered in oil
What are the fuel trims
What's the pre cat and post cat 02:readings
And so on.
I am going to guess, the fuel trims are maxed out
And the spark plugs are sooty
The pre cat is switching correctly between 0.2V and 0.8V
Due to a slug of unburnt air from one of the cylinder reaching the 02 sensor, and the ECU incorrectly adding more fuel
But this is a guess
It's easy to see which cylinder isn't firing you don't need any fancy equipment
Just unplug one injector at a time, the one that makes no difference is the cylinder that isn't firing
I use the HT leads and understand the dangers of bore wash, as long as you only run it for a few seconds it's quicker
Here's my panda running on 4,3 and 2 cylinders
Smooth on 4 cylinders, slight vibration on 3 and a pop pop sound especially from the tail pipe and very rough on 2