That’s quite far fetched.
> Chinese were first to rely on fermented grains to be food at least 6k years ago.
Not really. We found vessels that are that old, and that contained fermented grain, the rest is extrapolation. There is also evidence of some kind of beer making in Egypt not that long afterwards. Fermented things are present in almost any culture you could care to have a look at, even the remotest Amazonian tribes.
> solution? favor those who could consume it to avoid starvation, but would not touch it otherwise because it is so unpleasant (instant hangover).
6k years is awfully short for evolution to drive out a bit of genotype in a whole population. Also, again the same thing was happening in Egypt at around the same time, and we don’t have any proof that it did not happen elsewhere as well.
Also, a pet peeve of mine: evolution is not sentient. There is no being that “favor” or drive to extinction according to a pre-existing plan.
> Meanwhile brewing alcohol only started maybe 3k years ago in north western Europe so they are still figuring it out.
It does not work like that, there is no line in the sand for evolution to have worked instead of being “figuring it out”.
For some versions of a gene to be selected, you also need a driving force, in the form of either improved fitness in some environments, or the number of children. Alcohol causing a tiny increase in absolute terms of some specific types of cancers, that are not all that likely to kill someone before they reproduce, seems quite unlikely to be that driving force.
OTOH, alcohol is well known to increase social contacts, which is likely to result in more offsprings (see e.g. studies on alcohol consumption and teenage pregnancy rates).
Overall, this theory sounds truthy and consistent with a pop-culture understanding of evolution, but is not quite consistent with our understanding of history or biology.