some possible issues with dev config

[adf-ncgr]

per conversation in LIS-NCGR meeting today, I ran into a couple of things when trying to set up a dev instance to compile the modified code @alancleary sent me (for a feature not likely to make it into the main codebase in its present form); these may be worth trying to fix before the next release (although they are not blocking me at this time so not pressing):

• docker compose -f compose.yml -f compose.dev.yml --profile all up -d fails on apple M1, possible due to arm support being added to gprctools only in the last couple of days missing pre-compiled arm binary grpc/grpc-node#1405 (I have not verified that this upgrade will solve the issue, as I was having trouble getting an updated package spec to be accepted, though my memory is now fuzzy on those details)
• after switching to a linux-based VM, I got things to build, but wasn't able to get DOMAIN specified in an .env file to be respected by the dev server build (at least, such was my interpretation of the "Invalid Host Header" error I kept getting)
• after switching to just building the gcv client off the Docker file in the gcv repo (ie not using the compose method) things basically worked, except for a couple of things from the README.md in that repo that maybe need to be updated if we're still keeping that as an officially supported method ( one is that a port mapping needs to be supplied for the docker run command and the other is the the bind mount doesn't like -v .:/gcv , something about the number of characters in the volume spec not being acceptable; using pwd instead solves it)