OCT-1643: Teardown session after request finished

[kgarbacinski]

Description

My hypothesis is that the database scoped-sessions (that are implicitly created for each flask request) aren't closed properly. It results in the need for creating additional database connections to handle these "idle" sessions and then it makes the overflow. The idea is to make sure that we do db.session.remove() when every request is retrieved in our Flask application (as it comes out from SO threads and documentation, it doesn't happen automatically)

Definition of Done

1. Acceptance criteria are met.
2. PR is manually tested before the merge by developer(s).
   • Happy path is manually checked.
3. PR is manually tested by QA when their assistance is required (1).
   • Octant Areas & Test Cases are checked for impact and updated if required (2).
4. Unit tests are added unless there is a reason to omit them.
5. Automated tests are added when required.
6. The code is merged.
7. Tech documentation is added / updated, reviewed and approved (including mandatory approval by a code owner, should such exist for changed files).
   • BE: Swagger documentation is updated.
8. When required by QA:
   • Deployed to the relevant environment.
   • Passed system tests.

---

(1) Developer(s) in coordination with QA decide whether it's required. For small tickets introducing small changes QA assistance is most probably not required.

(2) Octant Areas & Test Cases.

[kgarbacinski]

/deploy