offload use of heavy `(load-commands)` and declare expected commands in `(:require)`

[gkspranger]

yes, major bikeshedding here, just trying to be more consistent and follow what kibit tells me

• src/yetibot/core/parser.clj

  • kibit told me to do this

• dropping heavy (load-commands) for something "lighter" and/or declaring tested bot commands in (:require)

  • test/yetibot/core/test/commands/cmd.clj

  • test/yetibot/core/test/commands/render.clj

  • test/yetibot/core/test/handler.clj

  • test/yetibot/core/test/interpreter.clj

  • test/yetibot/core/test/webapp/routes/graphql.clj

  • test/yetibot/core/test/parser.clj

  • no need for many uses of (load-ns) here, just use (:require)

[codecov[bot]]

Codecov Report

Merging #215 (d71e69e) into master (92434b0) will not change coverage. The diff coverage is 100.00%.

@@           Coverage Diff           @@
##           master     #215   +/-   ##
=======================================
  Coverage   62.58%   62.58%           
=======================================
  Files          95       95           
  Lines        4164     4164           
  Branches      233      233           
=======================================
  Hits         2606     2606           
  Misses       1325     1325           
  Partials      233      233           

Impacted Files	Coverage Δ
src/yetibot/core/parser.clj	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 92434b0...d71e69e. Read the comment docs.

[gkspranger]

i won't pretend to understand why the 1st run failed in relation to test.slack .. so then i tried to mimic how tests are ran locally, instead of just using lein test ..

so i ran tests with this command lein with-profile +test cloverage --codecov --runner midje -- and found a failure with test.chat -- one of those postgres yetibot_channel does not exist errors .. so i required test.db to ensure all DB tables would be created -- they were -- and everything seems to passed AOK ..

again, not entirely sure why test.slack failed the 1st time around -- but at least i now know how to mimic what is done via the GH actions ..