language_models: Improve token counting for providers (#32853)

We push the usage data whenever we receive it from the provider to make
sure the counting is correct after the turn has ended.

- [x] Ollama 
- [x] Copilot 
- [x] Mistral 
- [x] OpenRouter 
- [x] LMStudio

Put all the changes into a single PR open to move these to separate PR
if that makes the review and testing easier.

Release Notes:

- N/A

crates/language_models/src/provider/lmstudio.rs

@@ -7,7 +7,7 @@ use http_client::HttpClient;
 use language_model::{
     AuthenticateError, LanguageModelCompletionError, LanguageModelCompletionEvent,
     LanguageModelToolChoice, LanguageModelToolResultContent, LanguageModelToolUse, MessageContent,
-    StopReason,
+    StopReason, TokenUsage,
 };
 use language_model::{
     LanguageModel, LanguageModelId, LanguageModelName, LanguageModelProvider,
@@ -528,6 +528,15 @@ impl LmStudioEventMapper {
             }
         }
 
+        if let Some(usage) = event.usage {
+            events.push(Ok(LanguageModelCompletionEvent::UsageUpdate(TokenUsage {
+                input_tokens: usage.prompt_tokens,
+                output_tokens: usage.completion_tokens,
+                cache_creation_input_tokens: 0,
+                cache_read_input_tokens: 0,
+            })));
+        }
+
         match choice.finish_reason.as_deref() {
             Some("stop") => {
                 events.push(Ok(LanguageModelCompletionEvent::Stop(StopReason::EndTurn)));