Fixes #10739

Use the `gnrc_netif_t::pid` member instead of the pid of the current thread when generating the the `gnrc_netif_hdr` in `gnrc_netif_ieee802154::_recv` function.

Use the `gnrc_netif_t::pid` member instead of the pid of the current thread when generating the the `gnrc_netif_hdr` in `gnrc_netif_ethernet::_recv` function.

While the recursion in `gnrc_sixlowpan_frag` shouldn't be infinite we
still should avoid using recursions in general (also to be able to
statically analyze stack usage). This unrolls the recursion.

When having a non-6LN interface and a 6LN interface (e.g. on a border
router) the assertion can hit when a Router Advertisement is received.
This makes the check an `if` statement rather than an assertion, to
account for that case.

Co-authored-by: Gunar Schorcht <gunar@schorcht.net>

The `l2addr` and `l2addr_len` members of the `gnrc_netif_t` struct did
not show up in the rendered documentation.

Avoids confusion with generic CoAP and capitalization retains order
in documentation.

Otherwise, there will be leaks ;-).

Otherwise, there will be leaks ;-).

When issueing the sending of the next fragment the current version of
`gnrc_sixlowpan_frag` doesn't check if the queue is full. This leads to
leakage of the packet buffer, since when it is full, the package never
gets released.

This change adds a checks and error exits in case the queue is full.

The current implementation uses the assumption that the IPv6 address is
always based on an EUI-64, which it doesn't has to.

The RN2xx3 device has this parameter in milliseconds. So far, there is no
other device driver using loramac.h

Notice that RN2xx3 only has a get command for rxdelay2, not a set command.
Nevertheless, that parameter was changed to milliseconds as well.