Fix problem related to code optimization

The detection of a new parent has been moved in a call back of the S1
treatment. The previous implementation had some side effect due to the
fact that the detection was coming too late in some cases (pb with
erbest calculation)

The doRestart called by the servo is now extension specific. It help to
make a more cleaner restart.

The calculation of the erbest is no longer called if
ExternalPortConfiguration is enabled. The erbest is set when a new
foreign master is added.

The S1 treatment must be called only if the foreign master is
accepted/registered by the BMCA.

Bug description : When a MASTER port is at a lower port number than a
SLAVE port, the SLAVE port is jumping all the time between UNCALIBRATED
and SLAVE state.

When the PPS generation is forced by configuration, the PPS was disabled
when an instance was leaving the UNCALIB/SLAVE states to another state.

- When a new GM has been detected by the BMCA, the SLAVE state must
transition to UNCALIBRATED. If the state is already set to UNCALIBRATED
, nothing must be done.

- All announce messages not received from erbest are not been taken into
account. erbest is the best qualified foreign master.
- Comments added

Introduction of an other way to calculate the best foreign master. It
has been changed to fix bugs found during ISPC20119 (IOL test). Tests
showed unstable behavior in the computation of foreign master
qualification in some specific circumstances (time window of 8s with
announce messages every 5s)

Issue detected by IOL test in ICALEPS2019
The selection of the ErBest was not taking account all the time of the
qualification.

- New mechanism for clock degradation (degrade clock class, clock
accuracy, clock variance, timeSource, ptpTimeScale, frequencyTraceable,
timeTraceable)
- timeSource can be configured now
- If clock accuracy, clock variance and timeSource are not set in
ppsi.conf, default values are used depending of the clock class.

This field is used to store the active peer connected to an instance. In
wr_mon tool, it replaces the field peer on the display which was not
very explicit when many peers are connected on the same port. The field
activePeer takes the value of peer field in the following cases :

           MASTER          UNCALIBRATED/SLAVE
P2P     Rx PDELAY_RESP    Rx qualified ANNOUNCE
          (Filtered)
E2E       DELAY_RESP      Rx qualified ANNOUNCE

In other cases, the activePeer field is set to 0

The fields in timePropertiesDS are been reevaluated. The file bmc.c
contains a table showing the different use cases.

If a port is in INITIALIZING state, we must not leave from this state as
for a GM we must stay in this state until the GM is locked.

The frequencyTraceable and timeTraceable fields in timePropertiesDS are
now set to TRUE only when the defaultDS clock class is
PTP_CLASS_GM_LOCKED(6) or PTP_CLASS_GM_HOLDOVER(7). These field values
will follow the degradation of the clock class in case of locking issue.

Some default attribute values were not compliant with the PTP standard.
These default attribute values are used when :
- attributes are not defined in ppsi.conf file
- attribute values in ppsi.conf file are out of range

A master in P2P was responding to a Delay request message. A check for
the delay mechanism has been introduced to fix this issue.

Just comments, no code changes.

- Problem was coming from the interval to pico conversion when  the
interval is negative.
- fix also issue in wrs_ajust_in_progress

When the link state changes rapidly, it can't be detected by PPSi. This
is the case with the BTrain configuration. So we must detect that the
peer has changed and restart the servo for a slave and use the right
protocol (extension/ptp).

At startup the state of a instance is set to DISABLED and will be set
if the link is up to INITIALIZING. This is done to be coherent with the
curent behavior: the state becomes DISABLED when the link is down

The WR time is set in GM mode as follow :
1/ At startup the NTP time is set using wr_date script
2/ PPSi calls the tool wr_date with parameters "set host". It will be
called every time it detects a PLL transition from UNLOCKED to LOCKED
state. Also PPSi provide a error counter 'gmUnlockErr' which gives the
number of time the PLL unlocked.
3/ Parameters "-v set host" in wr_date tool, set only the second part of
the WR time. The time is set in a middle of two WR seconds ticks.

As the enable_output_timing() hook  is called also in the standard PTP
part, the hook has been moved from the WR specific structure
wrh_operations to  pp_time_operations

- Add a new state machine available for all timing modes ( Free running
master, Grand master and boundary clock)
- State is stored in pp_globals, field timingModeLockingState
- The 3 main states are :
-- Locking: Waiting for PLL locked
-- Locked: PLL is locked
-- Error: PLL is unlocked but was locked before
- Changing the timing mode will reset the state to 'Locking'

- Change the Quality structure to make it compliant with the standard
- Fix network to host conversion to display offsetScaledLogVariance

- Calculations made using fixed pointarythmetic
- delayAsymCoeff calculated using polynomial expansion
- Based on Rens Roosenstein work

- Simplify how an instance switch from WR to PTP protocol and vice-versa
- Change the WR state machine: When an unexpected WR signaling message
is received, the state is forced to IDLE. Solve handshake issue when
instance is waiting for an answer with a long time-out and in the other
side the PPSi process is restarted. In this use case the handshake will
fail.

- Has Hal is no longer used for ingress/egress latencies, we must then
take them from the configuration file
- Definition used by the common servo (WR & HA)
- Implementation of the IDLE state
See previous commit 'Review of WR protocol implementation'

In WR extension, the egress and ingress latencies are already included
in the exchanged deltas (TX/RX). This modification allows for an
extension to remove the egress/ingress latencies from the calculation of
the RX and TX timestamps. This feature is used in the WR extension.

Force to stay in UNCALIBRATED state if externalPortConfigurationEnabled
is set to true and desiredState to UNCALIBRATED.

- When the number of bit used for the calculation was smaller than 16,
the result of the calculation was 0.

- Removed master_msg, s1, execute slave

- For WR extension, we keep the old way to stay compliant with previous
version of WR
- The changes are applied to HA extension and compliant now with the
IEEE standard (delayAsymmetry sent in the CField of [P]DelayReq msg,
...)