Associate each thread with a model (#29573)

This PR makes it possible to use different LLM models in the agent
panels of two different projects, simultaneously. It also properly
restores a thread's original model when restoring it from the history,
rather than having it use the default model. As before, newly-created
threads will use the current default model.

Release Notes:

- Enabled different project windows to use different models in the agent
panel
- Enhanced the agent panel so that when revisiting old threads, their
original model will be used.

---------

Co-authored-by: Richard Feldman <oss@rtfeldman.com>

crates/language_model/src/language_model.rs

@@ -39,10 +39,14 @@ pub use crate::telemetry::*;
 pub const ZED_CLOUD_PROVIDER_ID: &str = "zed.dev";
 
 pub fn init(client: Arc<Client>, cx: &mut App) {
-    registry::init(cx);
+    init_settings(cx);
     RefreshLlmTokenListener::register(client.clone(), cx);
 }
 
+pub fn init_settings(cx: &mut App) {
+    registry::init(cx);
+}
+
 /// The availability of a [`LanguageModel`].
 #[derive(Debug, PartialEq, Eq, Clone, Copy)]
 pub enum LanguageModelAvailability {

crates/language_model/src/registry.rs

@@ -188,7 +188,7 @@ impl LanguageModelRegistry {
             .collect::<Vec<_>>();
     }
 
-    fn select_model(
+    pub fn select_model(
         &mut self,
         selected_model: &SelectedModel,
         cx: &mut Context<Self>,