Thanks Hedges, using the resistor brought down the voltage. I have cells close now at 3.592V and 3.570V, and another two
I'm going to top balance the other two which now sit at 3.342V and 3.338V. After all cells are close within range, I will set PSU to 3.65V and watch very closely ><
I believe my mistake was reconnecting all batteries in parallel and charging last night while 1 cell was already at 3.5V before the others. Other cells would have been at 3.3V, and due that magical SOC graph that I only intuitively understand, the higher charged cell would have charged more quickly?
As for the PSU jumping to 3.8V overnight, I'm not exactly sure. My explanation is that perhaps due to the lower quality of these benchtop supplies their voltages can jump over. Or perhaps it can be explained in the context of one cell charged higher than the others?
One thing to point out that I did notice last night was that when I attached the ring terminals at 3.65V on PSU, the power supply swapped over to CC instead of CV. I switched it off, disconnected and brought the PSU to 3.6V instead. Reconnected and returned it to CV. Wasn't sure if this was an issue, it was just something I hadn't encountered as of yet using these devices so wanted to be on the safe side.