Add thinking budget for Gemini custom models (#31251)

Closes #31243

As described in my issue, the [thinking
budget](https://ai.google.dev/gemini-api/docs/thinking) gets
automatically chosen by Gemini unless it is specifically set to
something. In order to have fast responses (inline assistant) I prefer
to set it to 0.

Release Notes:

- ai: Added `thinking` mode for custom Google models with configurable
token budget

---------

Co-authored-by: Ben Brandt <benjamin.j.brandt@gmail.com>

crates/google_ai/src/google_ai.rs

@@ -289,6 +289,22 @@ pub struct UsageMetadata {
     pub total_token_count: Option<usize>,
 }
 
+#[derive(Debug, Serialize, Deserialize)]
+#[serde(rename_all = "camelCase")]
+pub struct ThinkingConfig {
+    pub thinking_budget: u32,
+}
+
+#[cfg_attr(feature = "schemars", derive(schemars::JsonSchema))]
+#[derive(Copy, Clone, Debug, Default, Serialize, Deserialize, PartialEq, Eq)]
+pub enum GoogleModelMode {
+    #[default]
+    Default,
+    Thinking {
+        budget_tokens: Option<u32>,
+    },
+}
+
 #[derive(Debug, Deserialize, Serialize)]
 #[serde(rename_all = "camelCase")]
 pub struct GenerationConfig {
@@ -304,6 +320,8 @@ pub struct GenerationConfig {
     pub top_p: Option<f64>,
     #[serde(skip_serializing_if = "Option::is_none")]
     pub top_k: Option<usize>,
+    #[serde(skip_serializing_if = "Option::is_none")]
+    pub thinking_config: Option<ThinkingConfig>,
 }
 
 #[derive(Debug, Serialize, Deserialize)]
@@ -496,6 +514,8 @@ pub enum Model {
         /// The name displayed in the UI, such as in the assistant panel model dropdown menu.
         display_name: Option<String>,
         max_tokens: usize,
+        #[serde(default)]
+        mode: GoogleModelMode,
     },
 }
 
@@ -552,6 +572,21 @@ impl Model {
             Model::Custom { max_tokens, .. } => *max_tokens,
         }
     }
+
+    pub fn mode(&self) -> GoogleModelMode {
+        match self {
+            Self::Gemini15Pro
+            | Self::Gemini15Flash
+            | Self::Gemini20Pro
+            | Self::Gemini20Flash
+            | Self::Gemini20FlashThinking
+            | Self::Gemini20FlashLite
+            | Self::Gemini25ProExp0325
+            | Self::Gemini25ProPreview0325
+            | Self::Gemini25FlashPreview0417 => GoogleModelMode::Default,
+            Self::Custom { mode, .. } => *mode,
+        }
+    }
 }
 
 impl std::fmt::Display for Model {