feat(api): update moduleOffsets when loading/calibrating modules to get accurate labware position.

[vegano1]

Overview

We need to include the calibrated module offsets when calculating the labware position, this way that position is accurate for operations required by the system. This pr will hook in the calibrated data and load it from the disk if it exists, as well as update the existing calibration data when we calibrate a module.

Cloeses: RCORE-666

Test Plan

• Make sure we can still run the LoadModule protocol engine command
• Run the LoadModule PE command and make sure we are loading module offsets from disk and returning them in LoadModuleResult
• Run CalibrateModule PE command to get new calibration values, then run LoadModule PE command and make sure the vectorOffset value in LoadModuleResult returns the new calibrated value.

Changelog

• Load module calibration when creating protocol engine StateStore
• Handle CalibrateModule to update the module calibration value
• Add calibrated offset when calculating the module offsets in the modules.states.get_module_offset function

Review requests

• Is this the right approach?
• Am I missing something?

Risk assessment

• Low, flex work

Todo

• fix unit tests

[codecov]

Codecov Report

Merging #12484 (9aef193) into edge (44daef5) will decrease coverage by 0.01%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##             edge   #12484      +/-   ##
==========================================
- Coverage   73.78%   73.78%   -0.01%     
==========================================
  Files        2247     2247              
  Lines       61774    61771       -3     
  Branches     6452     6452              
==========================================
- Hits        45580    45577       -3     
  Misses      14674    14674              
  Partials     1520     1520              

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

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

Impacted Files	Coverage Δ
...pentrons/protocol_engine/create_protocol_engine.py	100.00% <ø> (ø)
.../protocol_engine/resources/module_data_provider.py	100.00% <ø> (ø)
api/src/opentrons/protocol_engine/state/state.py	100.00% <ø> (ø)
api/src/opentrons/protocol_engine/state/modules.py	90.81% <100.00%> (ø)

... and 1 file with indirect coverage changes

[sanni-t]

I think this way of including the offsets in loadLabwareResult and caching offsets after a calibrateModule command is quite a roundabout way of adding offsets. Also, unless I'm mistaken as to how module_id is used, it looks like you'll never be able to load a module offset from disk because the module_id generated by the engine is completely random and changes every run.
But before that, I have a basic question-
Do we intend to calibrate a module before each protocol run, like LPC, or is the plan to calibrate the module once, like an instrument calibration, save that info to disk, and then use those calibrated offsets each time that module is used in a protocol? Also, do we want to save unique offsets for a module+slot combo or do we want to use the same calibrated offset of a module for all slot locations?

[sanni-t]

Loading and applying offsets in engine looks good!