this allows users to adjust the base speed used for scaling roll/pitch
PIDs. This can be used to make PIDs work for both airspeed and
non-airspeed control

the SET_ROI commands are not supported yet

we are now 1.0 only

the old nav_roll will be used for the 2.50 release to prevent the need
for re-tuning. For the release after that we will use the new
calculation

Enable Mount config screen
Add Quick View (double click to change)
fix mono updater issue
RFD900 transmit values Mod
new mavlink controls (wip)

do_nav_roi and verify_nav_roi functions added to turn the copter and/or mount in response to a MAV_CMD_NAV_ROI way point

set_roi_cmd modified to take a Location.  I know the intention was for set_roi_cmd to interpret the MAVLink command directly but unfortunately in arduCopter missions are only made up of waypoints (which are Locations).

Keeps the failsafe from exiting Auto mode when returning from failsafe. Added arming ability for Toy mode.

When GA_b is small the direction of the vector is unreliable, so
normalising can exacerbate the error in the direction

Convert to IActivate, IDeactivate scheme, thanks andrew
add support for rfcomm* interfaces on linux
fix guage off screen draw mono issue.
remove use of BackStageViewContentPanel
andrews spacer changes - not using dues to screen space issue
change configpanel constructor to load xml directly
remove IMavlink Interface
fix hsi off screen draw issue on mono
modify hud to use sprite fonts, instead of drawing via GDI+
modify progress reporter to use a 10hz timer to update screen, using invoke/begininvoke fails on mono at 50hz (over 100ms per call).
fix targetalt and target airspeed jumping issue.
lots of cleanup on tab switching, ie stoping timers/other
3dr radio status led update
update ardurover car icon
speedup georef image screen. tested on over 1000 images.

when a user forces a waypoint jump, we may need to reset prev_WP to
prevent us thinking the waypoint is complete before we have started

the VCC pin number should not be converted

this solves a problem with yaw estimation when at high roll angles,
when the compass code becomes ineffective

no functionality changes