feat: Make max_tokens configurable via provider config

- Add max_tokens field to ModelProviderInfo struct
- Read from provider config, default to 8192 if not set
- Add debug logging for max_tokens value
- Allows per-provider customization in config.toml

Example config:
```toml
[model_providers.litellm]
max_tokens = 16384
```

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>

llmx-rs/core/src/chat_completions.rs

@@ -443,10 +443,12 @@ pub(crate) async fn stream_chat_completions(
     });
 
     // Add max_tokens - required by Anthropic Messages API
-    // Use a sensible default of 8192 if not configured
+    // Use provider config value or default to 8192
+    let max_tokens = provider.max_tokens.unwrap_or(8192);
     if let Some(obj) = payload.as_object_mut() {
-        obj.insert("max_tokens".to_string(), json!(8192));
+        obj.insert("max_tokens".to_string(), json!(max_tokens));
     }
+    debug!("Using max_tokens: {}", max_tokens);
 
     debug!(
         "POST to {}: {}",

llmx-rs/core/src/model_provider_info.rs

@@ -87,6 +87,10 @@ pub struct ModelProviderInfo {
     /// the connection as lost.
     pub stream_idle_timeout_ms: Option<u64>,
 
+    /// Maximum number of tokens to generate in the response. If not specified, defaults to 8192.
+    /// This is required by some providers (e.g., Anthropic via LiteLLM).
+    pub max_tokens: Option<i64>,
+
     /// Does this provider require an OpenAI API Key or ChatGPT login token? If true,
     /// user is presented with login screen on first run, and login preference and token/key
     /// are stored in auth.json. If false (which is the default), login screen is skipped,
@@ -290,6 +294,7 @@ pub fn built_in_model_providers() -> HashMap<String, ModelProviderInfo> {
                 request_max_retries: None,
                 stream_max_retries: None,
                 stream_idle_timeout_ms: None,
+                max_tokens: None,
                 requires_openai_auth: false,
             },
         ),
@@ -330,6 +335,7 @@ pub fn built_in_model_providers() -> HashMap<String, ModelProviderInfo> {
                 request_max_retries: None,
                 stream_max_retries: None,
                 stream_idle_timeout_ms: None,
+                max_tokens: None,
                 requires_openai_auth: true,
             },
         ),
@@ -375,6 +381,7 @@ pub fn create_oss_provider_with_base_url(base_url: &str) -> ModelProviderInfo {
         request_max_retries: None,
         stream_max_retries: None,
         stream_idle_timeout_ms: None,
+        max_tokens: None,
         requires_openai_auth: false,
     }
 }