fix(api): Flex 2 15 api fixed trash auto tipdrop support

[CaseyBatten]

Overview

Seeks to solve the API 2 15 support for RQA-1990

Test Plan

Run a 2.15 protocol on Flex that ends with a tip attached, ensure it auto drops in the fixed trash. Also run a 2.16 protocol on flex and ensure it does not auto drop them.
Run both on OT2 and ensure that the OT2 always drops tips.

Risk assessment

This should be fixing support for Flex, but has changed how we gate behavior exclusively for OT2, test thoroughly.

[codecov]

Codecov Report

Merging #14225 (dc4e765) into chore_release-7.1.0 (9842207) will decrease coverage by 1.86%.
Report is 6 commits behind head on chore_release-7.1.0.
The diff coverage is n/a.

Additional details and impacted files

@@                   Coverage Diff                   @@
##           chore_release-7.1.0   #14225      +/-   ##
=======================================================
- Coverage                70.39%   68.54%   -1.86%     
=======================================================
  Files                     2513     2514       +1     
  Lines                    71323    71416      +93     
  Branches                  9025     9061      +36     
=======================================================
- Hits                     50209    48949    -1260     
- Misses                   18908    20377    +1469     
+ Partials                  2206     2090     -116     

Flag	Coverage Δ
notify-server	89.13% <ø> (ø)

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

Files	Coverage Δ
...rons/protocol_engine/execution/hardware_stopper.py	100.00% <ø> (ø)
robot-server/robot_server/runs/engine_store.py	95.91% <ø> (ø)

... and 167 files with indirect coverage changes

[CaseyBatten]

Just ran tests with a protocol 2.15 and 2.16 on Flex and OT2, the 2.16 Flex protocol did not automatically drop tips (as expected). In all other situations it automatically dropped the tips, as expected.

[SyntaxColoring]

LGTM, thank you!

[SyntaxColoring]

I think this will work but it's pretty confusing.

For protocols that happen to load normal labware in slot A3, the way this will work now is get_fixed_trash_id() will return the ID of that normal labware, and then this code will ignore it because it doesn't match FIXED_TRASH_ID.

I think what we want instead is for get_fixed_trash_id() to either:

• Return the ID of the fixed trash labware, if there is one
  • And then this code would use that ID; this code wouldn't know about FIXED_TRASH_ID itself
• Otherwise, return the name of the fixed trash addressable area, if there is one
  • And then this code would use that addressable area name
• Otherwise, there must be no fixed trash, so it would return None