Refactors data store to sqlite

[lexicalunit]

It's ready!

TODO:

• [x] Refactor Data
• [x] Refactor Test Suite
• [x] Optimize
• [x] Migration

[lexicalunit]

Changes:

• Updates to the fish and bugs scripts to handle prices like "3,000" whereas before they were listed as just integers.
• The term "data" was being used for multiple things, so I changed it and now:
  • data is the stuff stored in the database
  • assets is the static application asset data
  • test/fixtures is the static test fixture data
• The migration script will leave behind the old .csv files just in case.
• I got rid of the configurable db_dir path as that could mess up migrations if people actually used it.
• You can collect rockin now and it knows that you mean Rockin' K.K.
• Adds a dependency on SQLAlchemy as it's the standard way to interface with databases in Python.
• The backup file that's created when a reset occurs is still a csv file, you can pretty easily import csv files directly into sqlite so that shouldn't be too much of an issue. The alternative would be to create a backup table in the database that could then be automatically rolled back to if there was an issue via some admin-only command like !rollback but that's future work.

What this gives us:

In the future we could now allow people to provide a database configuration and let them select the database they want to use. In development/tests we could use a sqlite database. Then in CI we could have a matrix that tests against sqlite and postgres. And in production people could optionally run against postgres. We'd need to have something like postgres in the future if we wanted to support multiple servers.

[lexicalunit]

We might also want to re-visit how we're using data.commit() in the future as wholesale overwriting the table data isn't what we want to have happen. Maybe pandas is smart enough not to be doing that 🤷‍♀️ In any case, we can probably refactor some of the commands at some point in the future to do things more optimally.

Edit: Fixed this!

[lexicalunit]

Updated PR to add Alembic database migrations so that we can easy migrate database over time.

I also added the first migration beyond the initial revision to add relationships to our data models. This could be super helpful in the future in refactoring things to be really nice. For example with relationship we can do things like:

user = get_user()
user.prices # get all this user's prices thru the relationship
user.fossils # get all this user's fossils thru the relationship
etc...

[codecov-io]

Codecov Report

Merging #171 into master will not change coverage. The diff coverage is 100.00%.

@@            Coverage Diff             @@
##            master      #171    +/-   ##
==========================================
  Coverage   100.00%   100.00%            
==========================================
  Files            2         4     +2     
  Lines          109       214   +105     
  Branches        13        12     -1     
==========================================
+ Hits           109       214   +105     

Impacted Files	Coverage Δ
src/turbot/assets.py	100.00% <100.00%> (ø)
src/turbot/data.py	100.00% <100.00%> (ø)
...rsions/versions/1afdca2a2389_adds_relationships.py	100.00% <100.00%> (ø)
...versions/versions/cb8aa87d74cb_initial_revision.py	100.00% <100.00%> (ø)

---

Continue to review full report at Codecov.

Legend - Click here to learn more Δ = absolute <relative> (impact), ø = not affected, ? = missing data Powered by Codecov. Last update 8e50568...3b13fc0. Read the comment docs.

[lexicalunit]

Also added docker documentation and fixed up some issues with #154.

[lexicalunit]

Added support for TURBOT_DB_URL so that we can theoretically use any database we want now.