Introduce a separate backend service for LLM calls (#15831)

This PR introduces a separate backend service for making LLM calls.

It exposes an HTTP interface that can be called by Zed clients. To call
these endpoints, the client must provide a `Bearer` token. These tokens
are issued/refreshed by the collab service over RPC.

We're adding this in a backwards-compatible way. Right now the access
tokens can only be minted for Zed staff, and calling this separate LLM
service is behind the `llm-service` feature flag (which is not
automatically enabled for Zed staff).

Release Notes:

- N/A

---------

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

crates/http_client/src/http_client.rs

@@ -175,6 +175,22 @@ impl HttpClientWithUrl {
             query,
         )?)
     }
+
+    /// Builds a Zed LLM URL using the given path.
+    pub fn build_zed_llm_url(&self, path: &str, query: &[(&str, &str)]) -> Result<Url> {
+        let base_url = self.base_url();
+        let base_api_url = match base_url.as_ref() {
+            "https://zed.dev" => "https://llm.zed.dev",
+            "https://staging.zed.dev" => "https://llm-staging.zed.dev",
+            "http://localhost:3000" => "http://localhost:8080",
+            other => other,
+        };
+
+        Ok(Url::parse_with_params(
+            &format!("{}{}", base_api_url, path),
+            query,
+        )?)
+    }
 }
 
 impl HttpClient for Arc<HttpClientWithUrl> {