Allow AI interactions to be proxied through Zed's server so you don't need an API key (#7367)

Co-authored-by: Antonio <antonio@zed.dev>

Resurrected this from some assistant work I did in Spring of 2023.
- [x] Resurrect streaming responses
- [x] Use streaming responses to enable AI via Zed's servers by default
(but preserve API key option for now)
- [x] Simplify protobuf
- [x] Proxy to OpenAI on zed.dev
- [x] Proxy to Gemini on zed.dev
- [x] Improve UX for switching between openAI and google models
- We current disallow cycling when setting a custom model, but we need a
better solution to keep OpenAI models available while testing the google
ones
- [x] Show remaining tokens correctly for Google models
- [x] Remove semantic index
- [x] Delete `ai` crate
- [x] Cloud front so we can ban abuse
- [x] Rate-limiting
- [x] Fix panic when using inline assistant
- [x] Double check the upgraded `AssistantSettings` are
backwards-compatible
- [x] Add hosted LLM interaction behind a `language-models` feature
flag.

Release Notes:

- We are temporarily removing the semantic index in order to redesign it
from scratch.

---------

Co-authored-by: Antonio <antonio@zed.dev>
Co-authored-by: Antonio Scandurra <me@as-cii.com>
Co-authored-by: Thorsten <thorsten@zed.dev>
Co-authored-by: Max <max@zed.dev>

crates/client/src/client.rs

@@ -13,7 +13,7 @@ use async_tungstenite::tungstenite::{
 use clock::SystemClock;
 use collections::HashMap;
 use futures::{
-    channel::oneshot, future::LocalBoxFuture, AsyncReadExt, FutureExt, SinkExt, StreamExt,
+    channel::oneshot, future::LocalBoxFuture, AsyncReadExt, FutureExt, SinkExt, Stream, StreamExt,
     TryFutureExt as _, TryStreamExt,
 };
 use gpui::{
@@ -36,7 +36,10 @@ use std::{
     future::Future,
     marker::PhantomData,
     path::PathBuf,
-    sync::{atomic::AtomicU64, Arc, Weak},
+    sync::{
+        atomic::{AtomicU64, Ordering},
+        Arc, Weak,
+    },
     time::{Duration, Instant},
 };
 use telemetry::Telemetry;
@@ -442,7 +445,7 @@ impl Client {
     }
 
     pub fn id(&self) -> u64 {
-        self.id.load(std::sync::atomic::Ordering::SeqCst)
+        self.id.load(Ordering::SeqCst)
     }
 
     pub fn http_client(&self) -> Arc<HttpClientWithUrl> {
@@ -450,7 +453,7 @@ impl Client {
     }
 
     pub fn set_id(&self, id: u64) -> &Self {
-        self.id.store(id, std::sync::atomic::Ordering::SeqCst);
+        self.id.store(id, Ordering::SeqCst);
         self
     }
 
@@ -1260,6 +1263,30 @@ impl Client {
             .map_ok(|envelope| envelope.payload)
     }
 
+    pub fn request_stream<T: RequestMessage>(
+        &self,
+        request: T,
+    ) -> impl Future<Output = Result<impl Stream<Item = Result<T::Response>>>> {
+        let client_id = self.id.load(Ordering::SeqCst);
+        log::debug!(
+            "rpc request start. client_id:{}. name:{}",
+            client_id,
+            T::NAME
+        );
+        let response = self
+            .connection_id()
+            .map(|conn_id| self.peer.request_stream(conn_id, request));
+        async move {
+            let response = response?.await;
+            log::debug!(
+                "rpc request finish. client_id:{}. name:{}",
+                client_id,
+                T::NAME
+            );
+            response
+        }
+    }
+
     pub fn request_envelope<T: RequestMessage>(
         &self,
         request: T,

crates/client/src/telemetry.rs

@@ -261,7 +261,7 @@ impl Telemetry {
         self: &Arc<Self>,
         conversation_id: Option<String>,
         kind: AssistantKind,
-        model: &str,
+        model: String,
     ) {
         let event = Event::Assistant(AssistantEvent {
             conversation_id,