I know this should probably be discussed on the Official Raspberry Foundation mailing list if such a thing exists, but a few thoughts:
1. At present, Raspbian enables ssh by creating a blank file called ssh in the boot partition and then booting the card. Could a similar method be used for enabling accessibility? 2. I'm pretty sure the lightest widespread option for getting speech at the console is espeakup or another front end to the speakup kernel module. Considering that Debian's talking installer defaults to installing and enabling espeakup when using the talking installer and not installing a GUI, it's probably not an unreasonable request to have it included in all official Raspbian images and enabling it on next boot if a file that's treated as an accessibility flag is present as suggested in 1. 3. Raspbian images that include a Desktop Environment probably should have Orca bundled by default with the standard keystroke for activating it on a system that doesn't autorun it. 4. Expanding on 1, perhaps the same access file can trigger making Orca autorun on images that boot into a Desktop environment and enable espeakup on images that only have the console. That said, unless things have changed significantly since I last setup a Pi, the only thing I had to do that I don't think would be needed if I was using the Pi as a sighted user was enabling ssh and using it to install and enable a text-mode screen reader, and I didn't have any trouble installing piespeakup following the directions on the Raspberry Vi website(though, admittedly, I used my PC to download, extract, and copy the tarballs to the Pi's SD card before loading the card into the Pi, and from there it was mostly running provided shell scripts). Haven't tried installing vanilla espeakup or Fenrir over ssh, so I can't comment on how hard they are to get up and running starting from a Raspbian Lite image.
