Integrate Method in Multi Session Integration lacks granularity

This is a common problem for me and no doubt others. 

Scenario - I have 5 sessions of data covering RGB or SHO. 

Say 25 Ha, 25 OIII but only 10 SII. 

I want to use Average integration for Ha and OII but Median for SII.

But I cant because the script forces me to choose one integration method for all frames. 

So what do I do?

Can we have a facility to allow the optimal Integration Method to be used for each stack please. 

UPDATE. Just done some testing to compare Median (No rejection) vs Average (Adaptive Rejection (Kappa 6-Low/4-High) and a bit surprised by the results for SII with only 10 frames.

In both cases the Average integration stacks were better quality - although differences were slight. I also noticed the rejection maps for Median showed noticeably more rejected data. Guess it could be data dependent but based on this I will now consider using Average  integration for >= 10 frames where previously I would have wanted >=20 frames.