Drunk people getting in fights is categorically different from violent crime committed for material gain (robbery, carjacking, etc). The causes for the problem and the way to go about solving the problem are totally different. Keeping people from getting angry drunk is not going to stop a home invasion. Stopping a home invasion is not going to improve crime stats because it's a single instance of violent crime. It's very possible to have a society relatively devoid of "violent crime" in which drunken brawls are fairly common. It's also possible to live in a society with lots of violent crime but no bar fights.
Including or excluding drunks to make the crime stats look how you want them to is no more honest than using gang violence to make the "mass shooting" statistics look how you want them to.
You can't just paint with a broad brush when it comes to violent crime (well you can but it's stupid and counterproductive if your goal is to understand crime for the purpose of advocating for public policy that reduces it). A drunk guy getting in a fight is different from domestic violence is different from robbing a delivery driver but they'll all show up when you "select * where includes_assault = true;"
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3015237/