The current name is confusing as it suggest an action,
which is switching to a new state. In reality these
functions first perform a check and then possibly
reject the state switch.
This also enables precision landing in landing mode, not just during
RTL.
This required an additional fix during the horizontal approach. It seems
like the z setpoint is not set when landing, so we can't blindly use
that setpoint but rather need to keep the altitude.
Currently only precision landing is using the
LandingTargetEstimator and it is a waste to have it
running in the background all the time. With this change
the estimator is only started once the precision landing
flight task is activated, and stopped when the flight
task is deactivated.
Alessandro Simovic