Move hook command execution to AgentSession.prompt()

Hook commands registered via pi.registerCommand() are now handled in
AgentSession.prompt() alongside file-based slash commands. This:

- Removes duplicate tryHandleHookCommand from interactive-mode and rpc-mode
- All modes (interactive, RPC, print) share the same command handling logic
- AgentSession._tryExecuteHookCommand() builds CommandContext using:
  - UI context from hookRunner (set by mode)
  - sessionManager, modelRegistry from AgentSession
  - sendMessage via sendHookMessage
  - exec via exported execCommand
- Handler returning string uses it as prompt, undefined returns early

Also:
- Export execCommand from hooks/runner.ts
- Add getUIContext() and getHasUI() to HookRunner
- Make HookRunner.emitError() public for error reporting

packages/coding-agent/src/core/hooks/index.ts

@@ -6,16 +6,18 @@ export {
 	loadHooks,
 	type SendMessageHandler,
 } from "./loader.js";
-export { type HookErrorListener, HookRunner } from "./runner.js";
+export { execCommand, type HookErrorListener, HookRunner } from "./runner.js";
 export { wrapToolsWithHooks, wrapToolWithHooks } from "./tool-wrapper.js";
 export type {
 	AgentEndEvent,
 	AgentStartEvent,
 	BashToolResultEvent,
+	CommandContext,
 	CustomMessageRenderer,
 	CustomMessageRenderOptions,
 	CustomToolResultEvent,
 	EditToolResultEvent,
+	ExecOptions,
 	ExecResult,
 	FindToolResultEvent,
 	GrepToolResultEvent,