Science go:
1. Zero point for the 24 hours of darkness, in fact categorical and objective harm will come from this (as well as possibly perpetuating bro-science).
2. The 1-2 weeks will be just barely long enough to see any impact of changing this piece of the environment.
3. The explicit amount of dark period in hours in going to be cultivar-dependent, a key fact of photoperiodism which is lost when folks hear the RULE OF THUMB that is 12/12 to flower. This is not the case for all strains, with 12 being a marginally safe estimate as a minimum dark period. Some flower at 0 hours of darkness, some flower at 10 hours, some flower at 14 hours per day.
4. Lux and PAR are not the same things and I really don't want to drive this point too hard considering you have a ppfd number but where did that come from? You need strict ppfd numbers, or a reeeeeeally good PAR map to do your DLI math. You are trading day hours for night hours, which is fine and you'll notice 100% a quicker flowering period at the cost of stress. The cost of those day hours needs to be accounted for with intensity changes on your light to hit your same DLI as if NOT a short day, right? Then and only then will you have anything of anything of anything regarding a mild idea of if CO2 doping will even help at all, if not harm. There is already demonstrable science on this point, some of which is on my page. Good thing that DLI in this sense is linear, and thusly you can use superposition to fudge these numbers much like engineers do for their daily maths, aka you'd double your light intensity (namely ppfd) to account for the 50% loss time.
Thanks for the research. Should do a diary and include the science you are compiling.