fix(RSpec split by examples): Properly determine slow test files when test example execution times and full test file execution time are both known

[ArturT]

Story

TODO: link to the internal story

Related

• improvement(RSpec): use test file execution times for existing test files on the disk to determine slow test files. This fixes issue with detecting slow test files when API token is shared between multiple test suites. #277

Description

About the problem:

We add up a spec file execution time and all test example execution times to determine the time of a test file, which doubles the execution time of a spec in the worst-case scenario. This leads to some tests not being split by test examples and unbalanced builds.

Changes

Use whichever execution time is larger:

• The execution time of the entire test file
• The sum of execution times of the test file examples

Checklist reminder

• You added the changes to the UNRELEASED section of the CHANGELOG.md, including the needed bump (ie, patch, minor, major)
• You follow the architecture outlined below for RSpec in Queue Mode, which is a work in progress (feel free to propose changes):
  • Pure: lib/knapsack_pro/pure/queue/rspec_pure.rb contains pure functions that are unit tested.
  • Extension: lib/knapsack_pro/extensions/rspec_extension.rb encapsulates calls to RSpec internals and is integration and e2e tested.
  • Runner: lib/knapsack_pro/runners/queue/rspec_runner.rb invokes the pure code and the extension to produce side effects, which are integration and e2e tested.