Merge branch 'stereocam2state_pullrequest_kim' of https://github.com/knmcguire/paparazzi into knmcguire-stereocam2state

* 'stereocam2state_pullrequest_kim' of https://github.com/knmcguire/paparazzi:
  [stereocam] New module for sending measurements from external stereocamera to state, with seperate conf files

[cleaning] remove lbeknav and fms folder

if libeknav should be used again, it can be included as a submodule from
https://bitbucket.org/jbrandmeyer/libeknav

Bebop i2c changes

Allow an external I2C bus connected through USB for the Bebop. Use the correct repeated-start for Transceive function for Linux targets as required by some I2C sensors.

We are changing/correcting the low-level I2C protocol for transceive
from START write STOP START read STOP
to START write START read STOP

the perl SimpleXML parser used for the tests can't handle it...

- only fold arrays on 'name' for aircraft, firmware and target to not get warnings about none unique 'name' attribs for modules and subsytems
- if airframe file could not be parsed, show error at the end

if libeknav should be used again, it can be included as a submodule from
  https://bitbucket.org/jbrandmeyer/libeknav

closes #1485

[stereocam] New module for sending measurements from external stereocamera to state, with seperate conf files

[gps] fix ubx ucenter autobaud and autoconf

The reason it is not working (at least one of the reasons...) is that the autobaud function is not working. Or exactly it is not very robust to properly catch the ACK. And if it happens to work (either it found the correct value while testing them or it falls back to the default and it is the correct one), it will not set the uart to the "new" value. It will send the message to set the baudrate to the last tested rate. This is because UBX_GPS_BAUD is defined to GPS_LINK.baudrate which is set to the last value, not the initial one.
The end result in most cases is that the modules are saturated with messages at 9600 bauds and that the reply of the version request fails, and then the NAV config is sent instead of NAV5.
I'm going to try this patch:

- save target baudrate at init
- use NAV5 by default if version couldn't be fetched

that split up parts of statements

closes #1454

[boards] naze32: PPM input on PA0 by default

That means by default you can't use SERVOS 1 and 2 and only the pwm channels on servo connectors 3-6 are active.

Chosing RADIO_CONTROL_PPM_PIN as PA7 (RC_CH6) via
<configure name="RADIO_CONTROL_PPM_PIN" value="PA7"/>
will also enable servos 1 and 2.

Rotorcraft guided mode

Set the autopilot mode to GUIDED and then send the GUIDED_SETPOINT_NED message to "goto" a position:
currently you can specify a frame in flags (first 4 bits):
- 0x0: LOCAL_NED, position in local NED frame
- 0x1: LOCAL_OFFSET_NED, position relative to where the vehicle currently is in NED
- 0x2: BODY_NED, same as LOCAL_NED for position
- 0x3: BODY_OFFSET_NED, position relative to vehicle pos AND heading (so X=forward, Y=right, Z=down)

Rate controller: sp in degrees/s

That means by default you can't use SERVOS 1 and 2.
Numbering is like on the silkscreen (use servos 3-6), like "standard mode" numbering of Abusemark naze32.

Chosing RADIO_CONTROL_PPM_PIN as PA7 (RC_CH6):
Enable servos 1 and 2.

meant for controlling the rotorcraft via external input (from a module or datalink message)

Currently positions mode only:
- specify frame via first 4 bits in flags:
  - 0x0: LOCAL_NED
  - 0x1: LOCAL_OFFSET_NED
  - 0x2: BODY_NED
  - 0x3: BODY_OFFSET_NED

Follow me based on stereo camera

Bebop2 motor order/spin direction changed for final version

should we really keep this xml_get.out program ?