#7771: Hooks: wire lifecycle events and tests

## Summary - Emit session_start/session_end hooks with resume info + message counts - Emit before/after compaction and gateway start/stop hook events - Wire message_sending/message_sent hooks for outbound delivery + reply dispatcher - Add unit tests for session and dispatcher hook payloads - Include tool hook runner context with toolCallId ## Testing - pnpm vitest run --config vitest.unit.config.ts src/auto-reply/reply/reply-dispatcher.hooks.test.ts src/auto-reply/reply/session.hooks.test.ts <!-- greptile_comment --> <h2>Greptile Overview</h2> <h3>Greptile Summary</h3> This PR wires additional plugin hook events across the agent lifecycle and outbound messaging pipeline. It adds a new unified `wrapToolWithHookRunner` to emit `before_tool_call`/`after_tool_call` (including `toolCallId`), emits `session_start`/`session_end` with resume info and message counts, emits compaction and gateway start/stop events, and adds unit tests validating dispatcher/session hook payloads. Main things to double-check are behavioral differences in error handling: some new hook invocations are fire-and-forget without `.catch()` (risking unhandled rejections), and the new tool wrapper changes behavior when `before_tool_call` throws (it currently blocks tool execution, whereas the previous wrapper continued). <h3>Confidence Score: 3/5</h3> - Moderately safe to merge, but there are a couple of behavioral/error-handling issues worth addressing first. - Most changes are additive hook emissions plus tests, but there are some non-trivial behavior changes: unhandled promise rejection risk from detached hook calls, and the new tool wrapper now fails tool execution if `before_tool_call` throws (likely unintended regression). - src/agents/pi-tools.hooks.ts, src/auto-reply/reply/session.ts, src/auto-reply/reply/agent-runner.ts, src/infra/outbound/deliver.ts, src/auto-reply/reply/reply-dispatcher.ts <!-- greptile_other_comments_section --> _{(3/5) Reply to the agent's comments like "Can you suggest a fix for this @greptileai?" or ask follow-up questions!} **Context used:** - Context from `dashboard` - CLAUDE.md ([source](https://app.greptile.com/review/custom-context?memory=fd949e91-5c3a-4ab5-90a1-cbe184fd6ce8)) - Context from `dashboard` - AGENTS.md ([source](https://app.greptile.com/review/custom-context?memory=0d0c8278-ef8e-4d6c-ab21-f5527e322f13)) <!-- /greptile_comment -->