Add support for getting the token count for all parts of Gemini generation requests (#29630)

* `CountTokensRequest` now takes a full `GenerateContentRequest` instead
of just content.

* Fixes use of `models/` prefix in `model` field of
`GenerateContentRequest`, since that's required for use in
`CountTokensRequest`. This didn't cause issues before because it was
always cleared and used in the path.

Release Notes:

- N/A

crates/language_models/src/provider/google.rs

@@ -344,9 +344,8 @@ impl LanguageModel for GoogleLanguageModel {
                 http_client.as_ref(),
                 &api_url,
                 &api_key,
-                &model_id,
                 google_ai::CountTokensRequest {
-                    contents: request.contents,
+                    generate_content_request: request,
                 },
             )
             .await?;
@@ -382,7 +381,7 @@ impl LanguageModel for GoogleLanguageModel {
 
 pub fn into_google(
     mut request: LanguageModelRequest,
-    model: String,
+    model_id: String,
 ) -> google_ai::GenerateContentRequest {
     fn map_content(content: Vec<MessageContent>) -> Vec<Part> {
         content
@@ -442,7 +441,7 @@ pub fn into_google(
     };
 
     google_ai::GenerateContentRequest {
-        model,
+        model: google_ai::ModelName { model_id },
         system_instruction: system_instructions,
         contents: request
             .messages