Fix build warnings across workspace

- Remove redundant token fields from StreamEvent::Finished (data
  already delivered via Usage event)
- Remove dead hotkey_adjust_sampling, MAX_HISTORY, now()
- Fix unused variable warnings (delta, log)
- Suppress deserialization-only field warnings (jsonrpc, role)
- Make start_stream/chat_completion_stream_temp pub(crate)
- Remove unnecessary pub(crate) re-export of internal types

Remaining warnings are TODO items: SkipIndex (scoring not wired),
notify (MCP notifications not wired).

Co-Authored-By: Proof of Concept <poc@bcachefs.org>

src/agent/api/mod.rs

@@ -13,8 +13,6 @@ mod openai;
 
 // Public API types — used outside agent::api
 pub use types::{Message, MessageContent, ContentPart, ImageUrl, Role, ToolCall, FunctionCall, Usage};
-// Internal types — re-exported for sibling modules within agent/
-pub(crate) use types::{ChatRequest, ReasoningConfig, ChatCompletionChunk, ChunkChoice, Delta, ToolCallDelta, FunctionCallDelta};
 
 use anyhow::Result;
 use std::time::{Duration, Instant};
@@ -72,8 +70,6 @@ pub(crate) enum StreamEvent {
     /// Stream finished.
     Finished {
         reason: String,
-        prompt_tokens: u32,
-        completion_tokens: u32,
     },
     /// Error from the stream.
     Error(String),
@@ -105,7 +101,7 @@ impl ApiClient {
     /// Start a streaming chat completion. Returns a receiver of StreamEvents.
     /// The caller (runner) reads events and handles routing to the UI.
     ///
-    pub fn start_stream(
+    pub(crate) fn start_stream(
         &self,
         messages: &[Message],
         tools: &[agent_tools::Tool],
@@ -137,7 +133,7 @@ impl ApiClient {
         (rx, AbortOnDrop(handle))
     }
 
-    pub async fn chat_completion_stream_temp(
+    pub(crate) async fn chat_completion_stream_temp(
         &self,
         messages: &[Message],
         tools: &[agent_tools::Tool],