I obviously have no idea how the math works out in detail, but I'd be pretty surprised if the economics of this were feasible. That is, spend a ton of money (and energy) to terraform sizable swaths of the Arizona desert just to avoid transporting in crops from Mexico right across the boarder? I'd be skeptical that even a back-of-the-napkin estimate would consider this possible. Relevant example: there are a bunch of rice farmers near Houston that are dependent on the Colorado River for irrigation (note, this is the Colorado River in Texas that runs through downtown Austin - completely different river from the Colorado River that goes through the Hoover Dam and supplies a ton of the Western US with water). Given how we've been getting drier over the past decades, the rice farmers are now frequently cut off from water because that water is deemed more important for city dwellers upstream where the economic return on that water usage is much greater.

If we can't even get enough water to these rice farmers (where it's actually relatively swampy, and note TX is a leader in renewable energy generation in the US), it seems like a silly pipe dream to talk about growing kale in the Arizona desert.