Refactor db

[javierggt]

Description

This PR fixes several issues having to do with the db.py module:

• Removes the "undo" feature which causes the file to bloat (#19)
• Removes the SQL backend (#20).
• Removes the use of filelock (#21).
• When saving, check that input tables have all the columns in the corresponding dtype (#23).
• Open the file in 'r' mode by default (#24).

It also fixes an issue that didn't have anything to do with db.py but it was convenient to do it at the same time:

• Do not calculate SNR, but use the value from celldetect (#22).

A few minor issues were found and fixed on the fly.

Fixes #19 Fixes #20 Fixes #21 Fixes #22 Fixes #23 Fixes #24

Interface impacts

This PR requires a new h5 file which can be found at ~jgonzale/tmp/astromon/astromon-pr-26.h5

The following columns were removed:

• astromon_obs
  • fids.
  • process_status.
• astromon_xray_src
  • name
  • double_id
  • status_id

Testing

Unit tests

• [ ] No unit tests
• [x] Mac
• [ ] Linux
• [ ] Windows

Independent check of unit tests by [REVIEWER NAME]

• [ ] [PLATFORM]:

Functional tests

I ran the same commands to create a new hdf5 file, with the code in this branch and the code in master. I then compared the output and verified that the differences are as expected, namely:

• the only column that is different is astromon_xray_src.snr
• fids and process_status columns were removed from astromon_obs
• name, double_id, and status_id columns were removed from astromon_xray_src

I ran the following commands:

   python -m astromon.scripts.get_cat_obs_data --workdir work --obsid 26266 --db-file pr-26.h5

   python -m astromon.scripts.get_cat_obs_data --workdir work --start 2022-01-04 --stop 2022-01-05 --db-file pr-26-2.h5

   python -m astromon.scripts.get_cat_obs_data --workdir work --obsid 26249 --db-file pr-26-3.h5

• For issue 24:

  In [1]: from astromon import get_cross_matches
  
  In [2]: xc = get_cross_matches(dbfile='astromon.h5')
  
  In [3]: !chmod -w astromon.h5
  
  In [4]: xc = get_cross_matches(dbfile='astromon.h5')

• For issue 19, I created a fresh file and made successive calls to db.save. I could clearly see the file bloating before this PR and it does not after.

[taldcroft]

@javierggt - I see in the functional testing that: "I intend to do the same tests as the previous PR". I'm not sure if you ended up doing that testing already, but with that testing completed this can be merged.

[javierggt]

Yes, I just edited the PR description to describe better what I did.