fix: preserve native ripgrep capabilities for direct models

Replace the restricted workspace_rg parameter mapping with native args and optional cwd. Preserve rg output and exit codes, and reuse process cancellation and output_read pagination so large search results remain accessible. Ask remains read-only within the workspace; Execute uses current-account permissions.

Validation: focused regression 169 passed, 1 skipped; typecheck and lint passed. Real DeepSeek Ask and Execute pagination scenarios passed with six counted requests in the successful run; an earlier failed fixture run has no retained request count. Full-suite validation is not green: the recorded run had unrelated failures and the latest run timed out without a summary. The remote Agent does not use this local execution path.

Release note: 直连模型可使用 ripgrep 原生参数搜索文件并保留原生结果，大输出可通过 output_read 完整续读。Ask 保持工作区只读，Execute 使用当前账号权限。

src/shared/builtin-model-tools.ts

@@ -18,7 +18,7 @@ export const builtinModelTools = [
     name: 'workspace_rg',
     displayName: '搜索工作区',
     description:
-      '使用 GoodBuddy 内置 ripgrep 搜索工作区内容或列出文件，返回带路径和行号的紧凑结果。',
+      '使用内置 ripgrep 原生参数搜索或列出文件；Ask 限于工作区只读搜索，Execute 使用当前账号权限，长输出可分页续读。',
     access: 'read',
     group: 'filesystem'
   },
@@ -113,7 +113,7 @@ export const builtinModelTools = [
   {
     name: 'output_read',
     displayName: '续读工具输出',
-    description: '按 cursor 分页读取当前会话已保留的进程或 Subagent 输出；会话释放后失效。',
+    description: '按 cursor 分页读取当前会话已保留的搜索、进程或 Subagent 输出；会话释放后失效。',
     access: 'read',
     group: 'programming'
   },