Naming of can_tx and can_rx about ros2_socketcan

[JWhitleyWork]

Yeah, I did copy this from kvaser_interface. However, the naming depends on whether you're looking at the data from the point of view of the ROS user or from the sensor. If you're coming from the sensor's point of view, sent would be data being sent from the sensor and received would be data received by the sensor. However, if you're looking at it from the point of view of the ROS user, sent would be data sent to the sensor while received would be data received from the sensor. While I agree that can_tx and can_rx might be backward, I still think they could be more precise.

To make it completely unambiguous, how do you feel about to_can and from_can or to_bus and from_bus?

[JWhitleyWork]

Changed to to_can_bus and from_can_bus in 468719f.

[kenji-miyake]

@JWhitleyWork I agree with you.
For the people who are familiar with CAN, tx and rx are fine, but others can't understand easily.
In addition, tx and rx cause the point-of-view problem, so I think there are better namings.

To make it completely unambiguous, how do you feel about to_can and from_can or to_bus and from_bus?

I think to_can and fron_can are ambiguous because the ROS message is can_msgs.
to_bus and from_bus (or to_can_bus and from_can_bus?) seem good to me!