KM3NET BROADCAST_BASE: Stay in WR_PRESENT state waiting for a targeted LOCK message and add some counters to test

concern DOMs (or other nodes) not concern the Bases. New additions will
come for Bases with BROADCAST_BASE statements.

KM3NET BROADCAST : - locking_enable in wr_link_on -> After unplug/plug or disable/enable the WR master port, the CLB stay locked in Uninitialized servo state because the spll stays locked.
		   - New WR state WRS_BROADCAST -> added for km3net monitoring via net.wr_st_gen variable.
		   - wr_servo_reset if errcount > 5 -> to play in a safe way.

KM3NET BROADCAST: We don't apply the HAS_FAULTS option with the broadcast configuration. We can enable the CONFIG_WRPC_FAULTS in parallel.

			If errcount > 5 set wrModeOn to False.

This modifications are needed because each time we reach an errcount > 5 (t1 timestamp flagged as
incorrect 5 consecutive times), we get stucked on uninitialized servo state.

KM3NET BROADCAST : Reset the timeout PP_TO_FAULT after receiving a follow-up message. Normally this is done after a response message, but we don't handle it.

KM3NET BROADCAST : set wrMode equal to WR_SLAVE in wr_handle_announce instead of wr_handshake_init (because in broadcast mode we never start the handshake).
		   + clear pp_lib_handle_announce.

KM3NET BROADCAST : Force the WRS_WR_LINK_ON after recive ANNOUNCE message in wr_handle_announce instead of pp_lib_handle_announce.

KM3NET BROADCAST : Reset the wr_servo if the TRACK_PHASE is not reached after a configurable number of iterations.

		   times t1 and t2 are used to compute the offset between the master and the slave cf
	           wr_e2e_offset_downlink().

		   doing handshake.

Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

abscal: simplify code related to CONFIG_ABSCAL
Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

For links longer than ~13km, a value (uint32_t) returned from a function
__div64_32 shifted by 16 bits overflowed, introducing PPS offset.

Bug introduced in the commit:
35eef679 time: introduce a new unified data structure
Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

For whatever reason we perform a normalization on an incorrect timestamp,
don't treat is as an negative value.
Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

This reverts commit a72f6bdc.

Remove warning that HAS_ABSCAL is redefined
Signed-off-by: Adam Wujek <adam.wujek@cern.ch>