chore(example/llm): Reintroduce models

See README

example/llm/README.md

@@ -4,3 +4,4 @@ While `services-flake` is generally used for running services in a *development*
 
 `example/llm` runs two processes ollama and open-webui, while storing the ollama data under `$HOME/.services-flake/ollama`. You can change this path in `flake.nix`.
 
+By default, a single model (`llama2-uncensored`) is downloaded. You can modify this in `flake.nix` as well.

example/llm/flake.nix

@@ -21,12 +21,20 @@
           # Backend service to perform inference on LLM models
           ollama."ollama1" = {
             enable = true;
+
             # The models are usually huge, downloading them in every project
             # directory can lead to a lot of duplication. Change here to a
             # directory where the Ollama models can be stored and shared across
             # projects.
             dataDir = "$HOME/.services-flake/ollama1";
+
+            # Define the models to download when our app starts
+            # 
+            # You can also initialize this to empty list, and download the
+            # models manually in the UI.
+            models = [ "llama2-uncensored" ];
           };
+
           # Get ChatGPT like UI, but open-source, with Open WebUI
           open-webui."open-webui1" = {
             enable = true;