Remove duplicate filenames from tables

[bhilbert4]

Resolves #26

In some cases, multiple copies of a given filename are present in the tables created from the MAST queries. This leads to repeated downloads of the same file. This PR remove the duplicate entires.

[bhilbert4]

Duplicates have been removed, but the product search does not make use of the starting and stopping time limits, so it is returning all files for the given propID.

• Need to find a way to associate start times with the rows in the product table, and then filter out entries that are outside of the user-given datetime bounaries.

[bhilbert4]

Looks like maybe we need to match up the parent_obsid from the products query with the obsid from the obs query. Then we can apply the t_min value from the obs query to the individual products from the product query, and filter appropriately.

Hmmm, that actually won't work. In the obsTable, all we have is the start time for the observation, not for individual files or even visits. So if the user requests an end time that is within the observation, we can't use the observation t_min to find the matching files.

[bhilbert4]

Tested the updated code using the example in #26. The number of files returned now matches what I find manually. Duplicates have been removed. Rather than trying to download a list of 27730 files, it now has a list of 158 uncal files plus 8 msa files.

[bhilbert4]

Merging this now, as it solves the problem of downloading multiple copies of files. Speeding up the query or making it smarter will come in a future PR.