Standardize on u64 for token counts (#32869)

Previously we were using a mix of `u32` and `usize`, e.g. `max_tokens:
usize, max_output_tokens: Option<u32>` in the same `struct`.

Although [tiktoken](https://github.com/openai/tiktoken) uses `usize`,
token counts should be consistent across targets (e.g. the same model
doesn't suddenly get a smaller context window if you're compiling for
wasm32), and these token counts could end up getting serialized using a
binary protocol, so `usize` is not the right choice for token counts.

I chose to standardize on `u64` over `u32` because we don't store many
of them (so the extra size should be insignificant) and future models
may exceed `u32::MAX` tokens.

Release Notes:

- N/A

crates/bedrock/src/bedrock.rs

@@ -152,7 +152,7 @@ pub enum Thinking {
 #[derive(Debug)]
 pub struct Request {
     pub model: String,
-    pub max_tokens: u32,
+    pub max_tokens: u64,
     pub messages: Vec<BedrockMessage>,
     pub tools: Option<BedrockToolConfig>,
     pub thinking: Option<Thinking>,

crates/bedrock/src/models.rs

@@ -99,10 +99,10 @@ pub enum Model {
     #[serde(rename = "custom")]
     Custom {
         name: String,
-        max_tokens: usize,
+        max_tokens: u64,
         /// The name displayed in the UI, such as in the assistant panel model dropdown menu.
         display_name: Option<String>,
-        max_output_tokens: Option<u32>,
+        max_output_tokens: Option<u64>,
         default_temperature: Option<f32>,
     },
 }
@@ -309,7 +309,7 @@ impl Model {
         }
     }
 
-    pub fn max_token_count(&self) -> usize {
+    pub fn max_token_count(&self) -> u64 {
         match self {
             Self::Claude3_5SonnetV2
             | Self::Claude3Opus
@@ -328,7 +328,7 @@ impl Model {
         }
     }
 
-    pub fn max_output_tokens(&self) -> u32 {
+    pub fn max_output_tokens(&self) -> u64 {
         match self {
             Self::Claude3Opus | Self::Claude3Sonnet | Self::Claude3_5Haiku => 4_096,
             Self::Claude3_7Sonnet