* peer_addr is now const
* new_link_udp_server() can bind to address or INADDR_ANY
* socket_udp_receive() happens without a loop
* socket_udp_send() happens without a loop
* sending in two packets (first size, then data) is only performed for TCP,
  SCTP, not for UDP anymore (it is unreliable, so we could miss something and
  will receive complete garbage)

RRCConnectionSetupComplete received in CU! (needed to allocate mem_pool for pdcp. remaining transactions on SRB1 and then SRB2 need to use PDCP (DCCH).

UL_RRC_MESSAGE_TRANSFER correctly received in CU. But : it shouldn't go directly to RRC, it has to go through PDCP via pdcp_data_ind

Merge branch 'feature-127-protocol-split' of https://gitlab.eurecom.fr/oai/openairinterface5g into feature-127-protocol-split

Conflicts:
	openair2/F1AP/f1ap_du_rrc_message_transfer.c

Merge branch 'feature-127-protocol-split' of https://gitlab.eurecom.fr/oai/openairinterface5g into feature-127-protocol-split

Merge branch 'feature-127-protocol-split' of https://gitlab.eurecom.fr/oai/openairinterface5g into feature-127-protocol-split

Conflicts:
	openair2/F1AP/f1ap_du_rrc_message_transfer.c

- add message sender (CU) + handler (CU) and fill with data from ITTI message
- add PLMNID_TO_MCC_MNC and BIT_STRING_TO_TRANSPORT_LAYER_ADDRESS_IPv4 macros
- correct TRANSPORT_LAYER_ADDRESS_TO_BIT_STRING:
  change name to TRANSPORT_LAYER_ADDRESS_IPv4_TO_BIT_STRING, make it send 4B, not 5

Fix for some variables

Merge branch 'feature-127-protocol-split' of https://gitlab.eurecom.fr/oai/openairinterface5g into feature-127-protocol-split

Conflicts:
	openair2/RRC/LTE/rrc_eNB.c

Merge branch 'feature-127-protocol-split' of https://gitlab.eurecom.fr/oai/openairinterface5g into feature-127-protocol-split

Conflicts:
	openair2/F1AP/f1ap_cu_rrc_message_transfer.c

Connection goes until generation of RRCConnectionSetup in CU which crashes (it calls rrc_mac_config_req. For DU this should be replaced with filling DL_RRC_MESSAGE_TRANSFER

modifications for SRB0 to allow multiple UE contexts in case of delayed CCCH reception in CU/DU split (moved SRB0 from carrier to UE context)

removed assertion in MAC layer and adjusted generation of MSG4 to handle when CCCH message is not present. UL CCCH is now received in CU, but asserts. Need to add code in DU to handle scheduling of DL-CCCH in later subframe. In current state, DL-CCCH will only come out if it is ready for MSG$ transmission.