[testbed/datasenders] add missing 0 to nanoseconds calculation

[pichlermarc]

Description:

Fixing a bug - the calculation when converting pdata timestamps to time is off in both the syslog and tcpudp datasenders.
Seconds are calculated correctly by dividing the timestamp by 10^9, but in nanosecond calculation the remainder of timestamp/10^8 is used, effectively always setting the first decimal in the resulting timestamp to 0.

Discovered this while writing some tests against the syslog receiver and receiving unexpected data in the log body, after identifying the issue a quick search revealed that the same thing was happening in the tcp/upd datasender.

Example:

2024-05-31T11:53:54.612287004+02:00 becomes
2024-05-31T11:53:54.012287004+02:00

Link to tracking Issue: no issue as it's a fairly minor thing that only affects tests, but can open if required

Testing:

• manual tests
• existing tests

Documentation: no additional documentation added.

[mx-psi]

Thank you!