* Fix several valgrind identified mem leaks

* Added callgrind target.

* px4_posix_tasks use nullptr

We need to protect access to the param_values array. This is dynamically
allocated and resized (utarray_reserve() calls realloc). If some thread
was iterating the array while another was resizing the array, the first one
would iterate on a freed array, thus accessing invalid memory.

On NuttX this could lead to hardfaults in rare conditions.

Unfortunately we need to initialize the semaphore on startup, by calling
sem_init(). This adds a param_init() method called by every board/config
that uses the params (at least I think I've found all of them)

This call is not needed, and will avoid deadlocks when locking is enabled.

This will ensure proper locking.

locking will be done in the params module

These changes were required to fix support for the ICM-20608 after this commit:

https://github.com/PX4/Firmware/commit/f746141afe3499ae56e75b3db06409381ead2f4b

This has been tested!

We buffer the mavlink messages and don't forward them directly
via uORB. This solves an issue where many requests are dropped and
QGC does not start properly in combination with UAVCAN devices.

Also change the initial state of the leds to off.

  * If you move in LAND, AUTO or HOLD the sticks the system will give control back to the pilot
  * If you do not connect any RC the system will default to HOLD and will allow you tablet control
  * If you gain position lock for the first time the system will re-evaluate the mode switch (so if you dropped down to alt hold it will now go into position)
  * If the system breaches the Geofence it will now always drop back to POSCTRL if the sticks are moved

We still want to be able to program the flight stack if the user changes
the baudrate of that UART.

We may need to change the baud rate from one version to the other.
Allowing the script to try multiple baud rates makes the transition a
little bit less painful.

This also fixes a bug in which it would go the next port before trying
to identify the board after asking it to reboot.

do same as command module

Since using more than 256 mission items has been tested and fixed, the
limit can now be raised to more than 256 (but smaller than UINT16_MAX).

Since the whole file is allocated in advance on NuttX, the limit there
is (arbitrarily) set to 2000. For POSIX archs such as SITL, we don't
really need to constrain ourselves any longer.

The dataman started overwriting contents after 256 items because the
item index were only uint8_t. This fix allows for more than 256
waypoints.