Change database systems

[latekvo]

Removing legacy TinyDB instances, they operate mainly in-memory, only ocassionally flushing to fs, which is highly undesireable.

Will be replacing them with SQLAlchemy, and likely hooking up PostgreSQL to it due to it's scalability.
We'll also need a document db, first candidate is of course Mongo but this is up for debate.

I think I'll also separate the downloaded documents out of the url_db and the summaries out of the summary_tasks db, into a pure document database.
This will allow us to have all the main databases to be Postgres, while using document based dbs only for pure text variable length data.

[latekvo]

This should be remembered, not sure what would be a good way of saving this info.
session.scalars(query).one() automatically raises an error when no results are found.

[latekvo]

Everything's ready, will test it better tomorrow, we'll have to dockerize postgres asap, that'll complete the entire thing i think.
Also, for now all the dbs are memory-kept, still SQL just in memory.

[latekvo]

summarizer is now working since the latest commit

[latekvo]

all workers are now launching 🥳

[Nekxis]

[latekvo]

(c.c. @SOSNE ) 🪨

[SOSNE]

Looks good to me