Describe the bug
When attempting to continue the conversation with an Ollama-based model, it will automatically fall back to the default model instead of continuing the conversation with Ollama.

Setup
Please complete the following information along with version numbers, if applicable.

• OS: macOS 14.4.1
• Shell: fish
• Terminal Emulator: WezTerm
• Terminal Multiplexer: WezTerm

To Reproduce
Steps to reproduce the behavior:

1. Type mods --setings
2. Insert:

apis:
  openai:
    base-url: https://api.openai.com/v1
    models:
      gpt-4:
        aliases: ["4"]
        max-input-chars: 24500
        fallback: gpt-3.5-turbo
  ollama:
    base-url: http://localhost:11434/v1
    api-key-env: NA
    models:
      "llama3:8b":
        aliases: ["llama3"]
        max-input-chars: 4000
default-model: gpt-4

3. install ollama
4. run ollama pull llama3:8b
5. type mods -M
6. choose llama3:8b
7. type whatever
8. run mods -C or mods -<SHA>
9. see that it is falling back to default GPT-4k
10. See error