Replace decimal year conversions with julian year conversions.

[sblunt]

See issue #372. This doesn't have any big impact on the Hipparcos IAD refitting, but maybe it affects things on longer timescales (i.e. hipparcos-gaia). Will continue to investigate.

[semaphoreP]

I think it generally looks good to me. Did you have a chance to run an HGCA orbit fit to see if it matches previous results?

[sblunt]

Ok, finally actually finished running a test fit, comparing with Hsu+ 2024 (case HGCA + no RVs). Here are my results:

sma1: 21.491 +4.059 / -3.620 au
ecc1: 0.212 +0.212 / -0.154
inc1: 43.180 +6.234 / -12.544 deg
aop1: 187.946 +111.395 / -117.375 deg
pan1: 199.073 +52.460 / -146.884 deg
tau1: 0.267 +0.200 / -0.128
plx: 37.896 +0.026 / -0.027 mas
m1: 47.1 +14.7 / -5.2 Mjup
m0: 1.110 +0.088 / -0.086 Msun

and Dino's results:

sma1: 19 +4 / -3
ecc1: 0.18 +0.19 / -0.13
inc1: 39 +7 / -14
plx: 37.9 +0.03 / -0.03
aop: 134 +202 / -113
pan: 54 +17 / -17
tau1: 0.4 +0.3 / -0.08
m1: 35 +5 / -3 Mjup
m0: 1.16 +0.09 / -0.09

They're surprisingly different! These are more similar to the orvara results (https://iopscience.iop.org/article/10.3847/1538-4357/ada3be), so I wonder if this is actually the root cause. I could do an apples-to-apples comparison fit with the orvara fit (including the same RVs) to be sure.