Unfortunately, lot's of whitespace changes, required to satisfy unspecified code style format errors.

Fixes shared memory locking bug and eliminates the need for an AppsProm driver to reserve a shared memory region.

Signed-off-by: tumbili <roman@px4.io>

Signed-off-by: tumbili <roman@px4.io>

If the failsafe state is ended when landed, we would switch back to
POSCTL and therefore take off again, however, all we want is stay on
ground and wait for the auto disarm.

The param COM_ARM_WO_GPS is set to 1 by default to allow arming without
GPS. This then sets a bool arm_without_gps which translates to
!GNSS_check in preflightCheck.

Previously, we renained in POSCTL and would drift away if GPS was lost.

The termination warnings were printed even if termination was not
actually enabled. This was confusing and is therefore fixed.

This reverts commit 9e704a5e121c7516e2133d02b328500d9d66fb67.

This reverts commit 77ea4cebf41cd106fe771b9eb469aa2326339467.

The deleted code conflicts with the failsafe handling in set_nav_state.
Also, flight termination was usually disabled by circuit breaker which
means this code had no effect anyway.

The param NAV_RCL_ACT was not implemented as described. Also, it has the
completely the wrong name. It should be a COM param and not NAV.
Therefore, remove the param and delete the partly implemented
and probably never used functionality.

We should definitely take action when GPS fails, this circuit breaker
shouldn't be engaged anymore.

If the navigator stops when no more position updates arrive, it won't
switch to land mode when DESCEND is requested by the commander.

The gps_failure flag had been ignored in some navigation states.

Always use the DESCEND mode and not LANDGPSFAIL because LANDGPSFAIL will
try to loiter for some time and then request termination instead of
descending gently and trying to land.

Implement code review feedback.  Add get_id_static to MavlinkStream items.  Add implementation of MAV_CMD_SET_MESSAGE_INTERVAL, MAV_CMD_GET_MESSAGE_INTERVAL.  Add deprecation message to REQUEST_DATA_STREAM.

MavlinkReceiver::handle_message_request_data_stream walks into deleted memory when you send the "stop" bit on a stream.  It also fails to restart the stream because it deletes the stream when you send the stop command, so restart needs to use stream_list to find the stream again.

Sometimes when flashing new firmware, the IO update fails because safety
is off. In this case, we should set safety on first before putting the
IO board into bootloader mode.