Planner: Tidied up the implementation of the parser and added a parameter meta data repository for easy access to the parsed data.

app.config: Added app.config key for the location of the files to be parsed for parameter meta data.

Parameters.pde: Updated comment schema to include the parameter name which becomes the XML node that contains the meta data.

fixup mode names in joystick and flight data
add mavlink 0.9/1.0 error message

useful for testing

this was preventing building MAVLink 1.0 with the arduino GUI as we
can't set cpp defines

most operations should now work

these are a very useful reference for the current headers in the code

this fixes the camera control code which was broken by a previous
import

this gives a more accurate picture of what is actually happening with
the servos

this allows logging of the actual servo output values. The radio_out
method previously used doesn't take account of the various override
mechanisms available via waypoints

Planner: Fixed flight mode binding issues. It is now configured to use the display text attribute as the text in the DDL and the integer value of the Enum as the value in the DDL. Parsing the selected value before saving to the board is unnecessary now because the value is already an int.

tidy up adams mod.
remove dup ch6_damp
rename level tab's
fixup long standing bug in flight planner field names
revert dtr order

ArduCopter/Navigation: Updated code to reference the new g.retro_loiter param instead of the compile define.

this buffers up _omega_I changes in _omega_I_sum over a period of 10
seconds, applying the slope limit only when _omega_I_sum is
transferred to _omega_I.

The result is a huge improvement in the ability of _omega_I to track
gyro drift over the long term.

the 'drop z' method reduced the impact of noise on omegaI, but it also
made us more sensitive to errors in accelerometer calibration and
scaling, as demonstated by the logs from Gabor here:

  http://diydrones.com/xn/detail/705844:Comment:834373

Simulation testing shows that the other noise suppression methods
applied in the DCM code, in particular the slope limiting on omegaI
the removal of the weighting and the upcoming use of a _omega_I_sum
buffer have reduced the impact of noise enough that we can now safely
include z in the acceleration calculation.