Extract completion provider crate (#14823)

We will soon need `semantic_index` to be able to use
`CompletionProvider`. This is currently impossible due to a cyclic crate
dependency, because `CompletionProvider` lives in the `assistant` crate,
which depends on `semantic_index`.

This PR breaks the dependency cycle by extracting two crates out of
`assistant`: `language_model` and `completion`.

Only one piece of logic changed: [this
code](https://github.com/zed-industries/zed/commit/922fcaf5a6076e56890373035b1065b13512546d#diff-3857b3707687a4d585f1200eec4c34a7a079eae8d303b4ce5b4fce46234ace9fR61-R69).
* As of https://github.com/zed-industries/zed/pull/13276, whenever we
ask a given completion provider for its available models, OpenAI
providers would go and ask the global assistant settings whether the
user had configured an `available_models` setting, and if so, return
that.
* This PR changes it so that instead of eagerly asking the assistant
settings for this info (the new crate must not depend on `assistant`, or
else the dependency cycle would be back), OpenAI completion providers
now store the user-configured settings as part of their struct, and
whenever the settings change, we update the provider.

In theory, this change should not change user-visible behavior...but
since it's the only change in this large PR that's more than just moving
code around, I'm mentioning it here in case there's an unexpected
regression in practice! (cc @amtoaer in case you'd like to try out this
branch and verify that the feature is still working the way you expect.)

Release Notes:

- N/A

---------

Co-authored-by: Marshall Bowers <elliott.codes@gmail.com>

crates/assistant/src/context.rs

@@ -1,12 +1,12 @@
 use crate::{
-    prompt_library::PromptStore, slash_command::SlashCommandLine, CompletionProvider,
-    LanguageModelRequest, LanguageModelRequestMessage, MessageId, MessageStatus, Role,
+    prompt_library::PromptStore, slash_command::SlashCommandLine, CompletionProvider, MessageId,
+    MessageStatus,
 };
 use anyhow::{anyhow, Context as _, Result};
 use assistant_slash_command::{
     SlashCommandOutput, SlashCommandOutputSection, SlashCommandRegistry,
 };
-use client::{proto, telemetry::Telemetry};
+use client::{self, proto, telemetry::Telemetry};
 use clock::ReplicaId;
 use collections::{HashMap, HashSet};
 use fs::Fs;
@@ -18,6 +18,8 @@ use gpui::{AppContext, Context as _, EventEmitter, Model, ModelContext, Subscrip
 use language::{
     AnchorRangeExt, Bias, Buffer, LanguageRegistry, OffsetRangeExt, ParseStatus, Point, ToOffset,
 };
+use language_model::LanguageModelRequestMessage;
+use language_model::{LanguageModelRequest, Role};
 use open_ai::Model as OpenAiModel;
 use paths::contexts_dir;
 use project::Project;
@@ -2477,9 +2479,10 @@ mod tests {
     use crate::{
         assistant_panel, prompt_library,
         slash_command::{active_command, file_command},
-        FakeCompletionProvider, MessageId,
+        MessageId,
     };
     use assistant_slash_command::{ArgumentCompletion, SlashCommand};
+    use completion::FakeCompletionProvider;
     use fs::FakeFs;
     use gpui::{AppContext, TestAppContext, WeakView};
     use indoc::indoc;