Revamp docker build process (#1121)

* refactor docker building

* change to buildx

* disable branch filter

* disable tags

* matrix for building

* fix branch filter

* rename workflow

* sanitize ref name

* fix sanitization

* fix source command

* fix source command

* add push arg

* enable for all branches

* logs

* empty commit

* try freeing disk space

* try disk clean again

* try alpine

* Update ghcr.yml

* Update ghcr.yml

* move checkout

* ignore .git

* add disk space debug

* add df h to build script

* remove pull

* try another failure bypass

* remove maximize build space step

* remove df -h debug

* add no-root

* multi-stage python build

* add ssh

* update readme

* remove references to config.toml

README.md

@@ -130,15 +130,13 @@ export LLM_API_KEY="sk-..."
 # The directory you want OpenDevin to modify. MUST be an absolute path!
 export WORKSPACE_DIR=$(pwd)/workspace
 
-docker build -t opendevin-app -f container/Dockerfile .
-
 docker run \
     -e LLM_API_KEY \
     -e WORKSPACE_MOUNT_PATH=$WORKSPACE_DIR \
     -v $WORKSPACE_DIR:/opt/workspace_base \
     -v /var/run/docker.sock:/var/run/docker.sock \
     -p 3000:3000 \
-    opendevin-app
+    ghcr.io/opendevin/opendevin:latest
 ```
 Replace `$(pwd)/workspace` with the path to the code you want OpenDevin to work with.
 
@@ -151,6 +149,9 @@ OpenDevin can work with any LLM backend.
 For a full list of the LM providers and models available, please consult the
 [litellm documentation](https://docs.litellm.ai/docs/providers).
 
+The `LLM_MODEL` environment variable controls which model is used in programmatic interactions,
+but choosing a model in the OpenDevin UI will override this setting.
+
 The following environment variables might be necessary for some LLMs:
 * `LLM_API_KEY`
 * `LLM_BASE_URL`