assistant: Stream tool uses as structured data (#17322)

This PR adjusts the approach we use to encoding tool uses in the
completion response to use a structured format rather than simply
injecting it into the response stream as text.

In #17170 we would encode the tool uses as XML and insert them as text.
This would require then re-parsing the tool uses out of the buffer in
order to use them.

The approach taken in this PR is to make `stream_completion` return a
stream of `LanguageModelCompletionEvent`s. Each of these events can be
either text, or a tool use.

A new `stream_completion_text` method has been added to `LanguageModel`
for scenarios where we only care about textual content (currently,
everywhere that isn't the Assistant context editor).

Release Notes:

- N/A

crates/assistant/src/terminal_inline_assistant.rs

@@ -1010,7 +1010,7 @@ impl Codegen {
         self.transaction = Some(TerminalTransaction::start(self.terminal.clone()));
         self.generation = cx.spawn(|this, mut cx| async move {
             let model_telemetry_id = model.telemetry_id();
-            let response = model.stream_completion(prompt, &cx).await;
+            let response = model.stream_completion_text(prompt, &cx).await;
             let generate = async {
                 let (mut hunks_tx, mut hunks_rx) = mpsc::channel(1);