Customize LLM config per agent (#2756)

Currently, OpenDevin uses a global singleton LLM config and a global singleton agent config. This PR allows customers to configure an LLM config for each agent. A hypothetically useful scenario is to use a cheaper LLM for repo exploration / code search, and a more powerful LLM to actually do the problem solving (CodeActAgent).

Partially solves #2075 (web GUI improvement is not the goal of this PR)

.github/workflows/dummy-agent-test.yml

@@ -31,7 +31,7 @@ jobs:
       - name: Run tests
         run: |
           set -e
-          poetry run python opendevin/core/main.py -t "do a flip" -m ollama/not-a-model -d ./workspace/ -c DummyAgent
+          poetry run python opendevin/core/main.py -t "do a flip" -d ./workspace/ -c DummyAgent
       - name: Check exit code
         run: |
           if [ $? -ne 0 ]; then