fix(rfi): remove automatic timestemp set and move default instrument to argument

[ljgray]

Setting the default timestamp to time.time() would potentially break compatibility with the cosmology daily pipeline since the current time does not necessarily have any relation to the CSD being processed.

For example, if an old day did not provide a timestamp, a band which has turned off since then would not get flagged when using the current time as a timestamp even though the band was still active when that data was collected.

I'm not sure if using the current time is needed for outrigger masking - if so, it should probably be moved to a separate place so that we maintain compatibility with the analysis pipeline.

[aelanman]

Thanks for fixing this, and apologies for introducing this error in the first place. I was getting an error before when the timestamp wasn't set, because (as I recall) in an earlier version it just did a comparison to the timestamp, and comparing a float to None raised an error.
The timestamp isn't currently used for outriggers, so it makes no difference there.

[ljgray]

Thanks! Originally it didn't support setting end times since I just assumed none of the bands would actually turn off... So the comparison would have been different and would have worked fine at that point using the current time.