Develop integration 2020 week 11

See merge request oai/openairinterface5g!787

Merge branch 'NR_RRCConfiguration' of https://gitlab.eurecom.fr/oai/openairinterface5g into NR_RRCConfiguration

Conflicts:
	executables/nr-softmodem.c
	executables/nr-ue.c
	executables/softmodem-common.c
	openair1/PHY/NR_TRANSPORT/nr_ulsch_decoding.c
	openair1/SCHED_NR/phy_procedures_nr_gNB.c
	openair1/SIMULATION/NR_PHY/dlsim.c
	openair1/SIMULATION/NR_PHY/ulsim.c
	openair2/LAYER2/NR_MAC_UE/mac_proto.h
	openair2/NR_PHY_INTERFACE/NR_IF_Module.c
	openair2/RRC/NR/MESSAGES/asn1_msg.c
	openair2/RRC/NR/MESSAGES/asn1_msg.h
	openair2/RRC/NR/nr_rrc_config.c
	openair2/RRC/NR/nr_rrc_defs.h
	openair2/RRC/NR/nr_rrc_proto.h
	openair2/RRC/NR/rrc_gNB.c

Fix for Issue #494: check and handle return value of openair0_device_load()  (develop branch)

See merge request oai/openairinterface5g!780

Fix l2 sim

See merge request oai/openairinterface5g!781

Signed-off-by: Raphael Defosseux <raphael.defosseux@eurecom.fr>

Actually there was a bug in FO compensation which I overlooked in last commit.

While fixing bug in PSS index calculation in 334519d16a9da34118c5568859eab6f68ae93ca9, I made changes in FO calculation as well. This is not nedeed and caused the FO to misbehave. Hence, reverting it.

This functions fills the ULSCH CQI indication message incorrectly. Also,
it should only be triggered when the eNB requests CSI. We disable it
here and a subsequent commit enables it again.

As adding PTRS leads to shortening of the coded data, redudandency is removed and the performance decreases on a pure AWGN channel.
So PTRS should only be enabled if phase noise is present.

NAS triggered the Kenb Refresh Req always for (hardcoded) UE 0. This
fixes this by setting the correct UE ID.

An effect of this was that the Kenb would always be set for UE 0,
regardless for which UE it was actually intended. This led for integrity
check failures for every UE with ID > 0.