I'm not sure I see the logic in your calculations for the minimum solar panel array Vmpp.
Firstly, I would say that it would be better to consider the actual MPPT controller output when it is providing absorption voltage to the battery, rather than the battery voltage itself.
For a 12V battery, the 12V 4S LiFePO4 Battery w/ BMS this is 14.5V.
Secondly, the 5 volts gap for the controller to turn on properly should be surely be in reference to the Open circuit voltage (Voc), not the Vmpp. If the MPPT is off, then no current can flow from the panels, and so they can only produce the Voc.
Thirdly, every source I've seen suggests that the voltage produced by solar panels remains more or less constant regardless of irradiance levels. This is true until only very low irrandiance is applied, at which point the panels are producing next to no power anyway. You can actually see a visualization of this in the charts at the bottom of the spec sheet posted by the OP.
Finally, I would say that its important to consider heating of the solar panel when looking at voltage. Most panels will have a voltage temperature coefficient of around -0.25%/C. The panel posted by OP, has (at STP) Vmpp = 38V, Voc temp coeff = -0.275% rated at a cell temperature of 25C. Lets suppose the panel heated up to 85C, the Vmpp should now fall to 38 * (1+(0.00275*60))=31.73. It would also be a good idea to consider voltage drop as the power flows from panel to SCC, lets assume a high value of 5% of this, reducing the minimum Vmpp the SCC would see from the panels to 30.13. This is more than enough for our 12V battery, and is actually sufficient for even a 24V setup. Victron MPPT controllers only require a 1V difference between the panels and the output power once they are turned on.