The unused bits are the lower bits, so we need to shift.

Not sure that the value 1 is what we want though.

The problem that may happen is that when the UE does not transmit
on PUCCH (for whatever reason) we may get no signal at all, and
thus compute a very low CQI. Later on we may ask the UE to transmit
louder which may lead to saturation and more problems.

The solution is simple: don't care about CQI in case of DTX
(and NACK also, as done for PUSCH).

Only FDD case done.
For TDD it seems that:
- nfapi structures are not correct. See nfapi_harq_indication_tdd_rel13_t
  in nfapi/open-nFAPI/nfapi/public_inc/nfapi_interface.h, all the cases
  (bundling, multiplex, ...) use nfapi_harq_indication_tdd_harq_data_t
- the function extract_harq does not handle TDD

There is too much UL scheduling because the buffer status is
overestimated.

This commit tries to make things simpler.

The following compilation did not work anymore:

    ./build_oai --eNB -w USRP -r Rel10 -c

(Rel8 still does not work.)

- flexran_get_half_duplex()
- flexran_get_intra_sf_hopping()
- flexran_get_type2_sb_1()
- flexran_get_ue_category()
- flexran_get_res_alloc_type1()

Through configuration, the lte-softmodem can be halted to await a
reconfiguration from the lte-softmodem. A mutex and condition variable are used
to implement this.

Furthermore, the dependance of lte-softmodem.c on FlexRAN has been reduced to
one include and the flexran_agent_start() call.

all the FlexRAN scheduling code is now also handled by the general OAI
scheduling code

- when changing values RC.rrc[i]->configuration should be changed too
- then, RRC_RECONFIGURATION_REQ will pass the changed conf down until PHY
- give warnings at functions that are not implemented

- rename functions
flexran_get_tdd_ack_nack_feedback() -> flexran_get_tdd_ack_nack_feedback_mode()
flexran_get_ue_pmi() -> flexran_get_ue_wpmi()