Refactor hook: split agent orchestration from formatting

- Remove POC_AGENT early return (was from old claude -p era)
- Split hook into run_agent_cycles() -> AgentCycleOutput (returns
  memory keys + reflection) and format_agent_output() (renders for
  Claude Code injection). poc-agent can call run_agent_cycles
  directly and handle output its own way.
- Fix UTF-8 panic in runner.rs display_buf slicing (floor_char_boundary)
- Add priority debug label to API requests
- Wire up F2 agents screen: live pid status, output files, hook log
  tail, arrow key navigation, Enter for log detail view

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

src/agent/runner.rs

@@ -301,6 +301,8 @@ impl Agent {
                                 // Flush display_buf except a tail that could be
                                 // a partial "<tool_call>" (10 chars).
                                 let safe = display_buf.len().saturating_sub(10);
+                                // Find a char boundary at or before safe
+                                let safe = display_buf.floor_char_boundary(safe);
                                 if safe > 0 {
                                     let flush = display_buf[..safe].to_string();
                                     display_buf = display_buf[safe..].to_string();