When building `nimble` with LLVM/clang the function
`ble_ll_adv_active_chanset_is_sec()` in `ble_ll_adv.c` is reported to
be unused, it is used within an assertions and `ifdef`s within that
file though. So this just disables that warning.

boards: slstk3402a: add i2c

kinetis: timer: #ifdef unused functions

Another issue revealed by compiling several apps with LLVM/clang for
the `teensy31` board.

cpu/nrf51: fix formatting uint to PRIu32

Release notes for 2018.07

This release-notes where contributed by the following github users:

* aabadie
* bergzand
* cladmi
* gebart
* miri64
* PeterKietzmann

pkg: update ccn-lite and adapt shell commands

gnrc_netif_ieee802154: Set FCF Frame Pending when more data is expected

gnrc_netif: Introduce GNRC_NETIF_HDR_FLAGS_MORE_DATA

Set frame pending bit in IEEE 802.15.4 FCF when
GNRC_NETIF_HDR_FLAGS_MORE_DATA is set on the netif header on outbound
frames.

cc2538: cleanup and optimisation of periph timer

cortexm_common/ldscript: simplify generation of firmwares in section of the ROM

tests/gnrc_netif: Fix -Wformat-overflow

Explain the test and the output you should get.

Verify that specifying a too big _fw_rom_length for the rom is detected and
prevent compilation.

Compile an elf file with a length equals to half the rom length.

Allow defining a specific rom length to use for linking the firmware,
_fw_rom_length, instead of the default configuration to use the whole rom from
_rom_offset to the end.

 * Add cortexm_common/Makefile.include FW_ROM_SIZE configuration
 * Add an assertion that _fw_rom_length still respects _rom_length

Compile two elf files with different offset and verify the linked file offset.
I only enabled samr21-xpro and iotlab nodes for the moment.

Define _rom_offset with a conditional evaluated at execution time to allow
setting it in compilation rules and generate in the same make instance different
elf files with different configurations.

The variables should all always be defined.

Test that _rom_offfset is removed from the available _rom_length.

It will help testing if it is taken into account and for defining for outside
after.

Trigger an overflow by 1 byte to detect in ROM_LEN is indeed used.

Inspired by kaspar030 version to removing the new _boot_offset variable.

https://github.com/kaspar030/RIOT/blob/cbf324a66d0c925dd02eca290e17c350945e97b2/cpu/cortexm_common/ldscripts/cortexm.ld

Fixes the following error with GCC 7

main.c: In function 'test_netif_get_name':
main.c:1054:23: error: 'sprintf' may write a terminating nul past the end of the destination [-Werror=format-overflow=]
     sprintf(exp_name, "if%d", (int)netif);
                       ^~~~~~
main.c:1054:5: note: 'sprintf' output between 4 and 9 bytes into a destination of size 8
     sprintf(exp_name, "if%d", (int)netif);
     ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

 Makefile.include: Include VERSION file for release

When downloading the release archive and building an example, the RIOT_VERSION string is not set to "Version 2018.04" but to
    
    Version: UNKNOWN (builddir: /home/me/Downloads/RIOT-2018.04)

This allows sourcing a global VERSION file to manually set it before releases.

make: introduce common Python lib path

boards*: documentation cleanup

- Cleanup brief description to be consistent with other boards
- Move general documentation in doc.txt

- Cleanup brief description to be consistent with other boards
- Move general documentation in doc.txt

Display brief description instead of link to the wiki

Display brief description instead of copyright

- display brief description instead of copyright
- add a reference to the 'real' documentation, otherwise it's empty

cpu/cc2538: refine periph/spi implementation

Introduce dist/pythonlibs directory to store RIOT python packages.
This directory is exported via PYTHONPATH by the build system to
make it commonly available.