PID 0, ppid(), parent() inconsistencies

[giampaolo]

On OSX:

>>> psutil.pids()[0]  # lowest pid
0
>>> psutil.Process(0)
psutil.Process(pid=0, name='kernel_task', started='2019-02-24 04:41:17')
>>> psutil.Process(0).ppid()
0
>>> psutil.Process(0).parent()
psutil.Process(pid=0, name='kernel_task', started='2019-02-24 04:41:17')
>>> 

On Linux:

>>> psutil.pids()[0]  # lowest pid
1
>>> psutil.Process(1)
psutil.Process(pid=1, name='systemd', started='2019-02-22 10:56:55')
>>> psutil.Process(1).ppid() 
0
>>> psutil.Process(1).parent()
>>> psutil.Process(0)
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "psutil/__init__.py", line 444, in __init__
    self._init(pid)
  File "psutil/__init__.py", line 484, in _init
    raise NoSuchProcess(pid, None, msg)
psutil.NoSuchProcess: psutil.NoSuchProcess no process found with pid 0

On OSX, ppid() return 0, which is "weird", but also understandable because it's the only possible "null" value. On the other hand parent() should return None (and this should be fixed for sure).

On Linux the "weird" thing is that ppid() return 0 (this comes from the kernel), but PID 0 does not exist. Perhaps it should return 1 instead.

[giampaolo]

Probably a reasonable solution here is to make sure that parent() always return None when dealing with PID 0 (or 1 on Linux) and leave ppid() alone.
ppid() should keep returning 0 also on Linux, even if PID 0 does not exist, because it's the only possible "null" value.

[giampaolo]

Fixed in 682fae3.