[Bug]: ECI conversions give wrong result for sub-second dates (juliandate issue)

[Pokornz]

What happened?

When trying to convert to/from ECI coordinates with dates that don't have whole seconds (i.e. milliseconds are not zero), the sub-second information is ignored in the Julian date conversion: https://github.com/geospace-code/pymap3d/blob/main/src/pymap3d/sidereal.py#L87
When used to for ex. convert satellite trajectories, the resulting trajectory has stair-like shape.

Relevant log output

No response

[Pokornz]

It can be pretty much fixed by adding the microseconds
C = (((time.second + time.microsecond / 1e6) / 60.0 + time.minute) / 60.0 + time.hour) / 24.0

[Pokornz]

I accidentally closed the issue.
I have now submitted a pull request. It is my first time contributing to a code base, so any feedback is welcome.
@scivision does this change brake anything/ is there a reason why sub-second information is not used?

[scivision]

fixed by #85