chore(mypy): fix cache cleanup

[ahiuchingau]

Overview

We haven't been properly clearing the mypy cache for some python directories when we do make clean-py because we were using the wrong paths.

It seems like we don't clear pytest caches in most directories in the cleanup rules, is that intentional?

[sfoster1]

Do we have to? It does tend to slow things down. Maybe we can make clean-extra targets if we really want to?

[ahiuchingau]

Do we have to? It does tend to slow things down. Maybe we can make clean-extra targets if we really want to?

Do you mean the pytest cache or mypy or both? Do you propose that I move the mypy cache and add pytest cache to a different target (makes sense and I'm very ok with that)?

For context, I noticed this only because I was switching between branches that have different pipenv requirements and for some reason mypy was not behaving unless I manually removed the cache first. So I thought not clearing the cache properly was the problem. But now I can't seem to reproduce the same issue I was having so I'm stumped.

[sfoster1]

Do we have to? It does tend to slow things down. Maybe we can make clean-extra targets if we really want to?

Do you mean the pytest cache or mypy or both? Do you propose that I move the mypy cache and add pytest cache to a different target (makes sense and I'm very ok with that)?

For context, I noticed this only because I was switching between branches that have different pipenv requirements and for some reason mypy was not behaving unless I manually removed the cache first. So I thought not clearing the cache properly was the problem. But now I can't seem to reproduce the same issue I was having so I'm stumped.

Yeah the thing is that running mypy over and over with no cache is slower, and we always clean when we build, so you'd end up with slow behavior when you do normal dev processes.

That said we definitely shouldn't have wrong paths in there, i wonder if having like clean-build removing build/ and such and clean: removing all caches would be a good idea. then the build targets depend on clean-build and clean is only ever run by humans.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 92.43%. Comparing base (8f50b08) to head (f0bceb9).
Report is 234 commits behind head on edge.

Additional details and impacted files

@@             Coverage Diff             @@
##             edge   #15072       +/-   ##
===========================================
+ Coverage   67.50%   92.43%   +24.93%     
===========================================
  Files        2521       77     -2444     
  Lines       72090     1283    -70807     
  Branches     9311        0     -9311     
===========================================
- Hits        48666     1186    -47480     
+ Misses      21228       97    -21131     
+ Partials     2196        0     -2196     

Flag	Coverage Δ
api	?
app	?
components	?
g-code-testing	92.43% <ø> (ø)
hardware	?
hardware-testing	?
labware-library	?
notify-server	?
ot3-gravimetric-test	?
protocol-designer	?
react-api-client	?
robot-server	?
shared-data	?
step-generation	?
system-server	?
update-server	?
usb-bridge	?

Flags with carried forward coverage won't be shown. Click here to find out more.

see 2444 files with indirect coverage changes

[sfoster1]

looks good, ty!