Going back to fixed shift in log2_approx for TM4. Playing with computing the channel average in order to find optimal shift.

2.Fixing shift LUT for TM1. Mcs 21, 26 and 28 show erroneous performance. (see attachement in commit)

to be completed in the next commit.

Introducing different shifts for channel compensation log2_maxh0 and log2_maxh1 as well as mcs-dependent offset coefficients interf_unaw_shift_mcs
for I-UA receiver and interference-free case TM4. Coeeficients for 64 QAM must be revised.

Introducing an offset for I-UA receiver intefr_unaw_shift and calibrating this value for each mcs (u=0 only).

Small changes in dlsch_demodulation.c and in dlsim.c condition to stop simulation depending on rx_type.
dr_offset must be calibrated for tm4 again.

Please note that to test precoder it is necessary to run Perfect_CE

Conflicts:
	openair1/SIMULATION/LTE_PHY/dlsim.c
	openair1/SIMULATION/TOOLS/random_channel.c

1. Fixing precoder selection criteria for TM4 for tpmi2.
TPMI 2 works now.
2. Temporarily replacing abstraction.c with an old version till fix the bug in init_freq_channel.
3. Commenting some extra printouts.

Merge branch 'feature-59-tm4' of https://gitlab.eurecom.fr/oai/openairinterface5g into feature-59-tm4