fix(computer_use): correct type_text MCP tool name and implement drag action

Bug 3: The cua_backend type_text() method called MCP tool 'type_text_chars'
which does not exist in current cua-driver. Changed to 'type_text' which is
the correct MCP tool name.

Bug 4: The drag() method returned a hardcoded 'not supported' error even
though cua-driver exposes a 'drag' MCP tool. Implemented proper drag
dispatching with coordinate-based and element-based targeting.

Added dispatch-level validation for drag to ensure from/to coordinates
or elements are provided before calling any backend.

Fixes #24170 (bugs 3 and 4)

tools/computer_use/tool.py

@@ -357,6 +357,12 @@ def _dispatch(backend: ComputerUseBackend, action: str, args: Dict[str, Any]) ->
         return _maybe_follow_capture(backend, res, capture_after)
 
     if action == "drag":
+        has_elements = args.get("from_element") is not None and args.get("to_element") is not None
+        has_coords = args.get("from_coordinate") and args.get("to_coordinate")
+        if not has_elements and not has_coords:
+            return json.dumps({
+                "error": "drag requires from_coordinate/to_coordinate or from_element/to_element",
+            })
         res = backend.drag(
             from_element=args.get("from_element"),
             to_element=args.get("to_element"),