Possible issue with Euler angle calculation (Yaw and roll).

[mzhouURI]

Describe the bug
When setting use_enu_frame to true, we discover the Yaw angle will decrease when placed sensor with z-axis pointing downward and rotate the sensor counter-clockwise.
I also found in this placement, the roll angle is reporting zero. but in ENU should it report 180 deg? Or if roll is reporting 0 deg the counter clockwise rotation should result in yaw angle increase.

To Reproduce
Steps to reproduce the behavior:

1. Launch Node with command roslaunch microstrain_inertial_driver microstrain.launch

Expected behavior

• Visualize yaw angle using plotjuggler.
• place sensor with z-axis pointing down and turn it counter clockwise, yaw angle will decrease, roll report 0 deg
• place sensor with z-axis pointing up and turn it counter clockwise, yaw angle will increase, roll report 180deg

Environment (please complete the following information):

• OS: ubuntu20.04
• Architecture: x86_64
• ROS Version: noetic
• Version: installed using sudo apt install ros-noetic-microstrain-inertial-drive
• Sensor(s): 3DM-GX5-25

Modifications
I believe is the Euler angle calculation may be wrong?

Launch Parameters
standard param used in the launch file.

Additional context