Hi Scott,
I'm looking for the same functionality. If that's OK I'd like to answer your
question.
 If we could see the source/origin docker-compose YAML, 
The Yaml might get messed up so you can also get it from here:
https://gist.github.com/rudolfvesely/4ad6f762eafbe5417f80ff469b690c38
I intentionally made it complicated to show various options from
"docker-compose".
- "build: ." will use Dockerfile (in case of Podman this should be
Containerfile) to build image. In other words this will run "docker build ."
- "args" will define / overwrite values in "ARG" in
Dockerfile/Containerfile
- "dockerfile: ./foo/bar/containerfile_for_db" is used when Dockerfile has
different name. In other words this will run "docker build
--file=foo/bar/containerfile_for_db"
$ cat ./something/api.env
 NODE_ENV=test 
$ cat .env
 TAG=v1.5 
$ cat docker-compose.yml
 version: "3.9"
 services:
   webserver:
     # "Dockerfile" is located in the working directory and will be used to
build image
     build: .
     # build arguments (used in Dockerfile)
     args:
       APP_HOME: app
     # environment variables
     environment:
       FLASK_ENV: development
   dbserver:
     # "Dockerfile" has different name
     dockerfile: ./foo/bar/containerfile_for_db
     # environment variables in a file
     env_file:
      - ./something/api.env
   redis:
     # variable defined in ".env"
     image: "webapp:${TAG}"
     volumes:
       - logvolume01:/var/log
     ports:
       - "5000:5000" 
Please let me know if you have other questions.
Kind regards,
Rudolf Vesely