fix(api, app): Clear instrument offset before performing deck calibration

[Laura-Danielle]

Overview

Closes #5022 by resetting instrument offset for all pipettes any time a new deck calibration begins. Some design changes were also added in the app by following instructions from this design.

Changelog

• Add messaging in deck cal CLI about re-doing pipette calibration (instrument offset does not get taken into account here because robot settings/deck cal are reset before beginning the program)
• Modify the app according to the design above
• The robot no longer restarts after a deck calibration
• Reset instrument offset in endpoints.py and reload the current pip instrument offset to be (0, 0, 0)
• Remove feature flags for v1 in endpoints.py as it's no longer relevant

Review requests

Please check that the code place for resetting instrument offset etc are OK.

Test Plan

1. Verify the issue with the robot and app on 3.19 or edge
   a. Modify your instrument offset (the value in /data/robot_settings.json under instrument_offset and the mount and type of pipette you'll use for deck calibration - the output of tip probe) to be bad, for instance having a value of 20 in its z component.
   b. Run deck calibration, trying to do a good job.
   c. verify that after the robot restarts, the deck calibration is obviously incorrect - if you modified the z component of instrument offset, the deck calibration z offset should be very wrong
2. Verify that this PR fixes the issue, after putting this PR on the robot with make push and building the app on this pr with make -C app dev
   a. Edit /data/robot_settings.json and verify that the value under instrument_offset for the mount and type of pipette used for deck calibration is still bad and large
   b. Run deck calibration, trying to do a good job
   c. When ending deck calibration, the button should say "save and exit" rather than "save and restart robot". The robot should not restart when you click the button
   d. Edit /data/robot_settings.json and verify that the instrument_offset for the mount and type of pipette you did deck calibration with is now (0, 0, 0)
   e. Edit /data/gantry_calibration.json and verify that the deck calibration is not bad in the same way it was in 1c
   f. Run the calibrate-to-crosses protocol used by the labware creator. Run tip probe and check that the pipettes move to the crosses accurately (or as accurately as they normally do).

Let me know if you have any other questions!

Risk assessment

Medium. There was concern that customers would be confused by the new messaging at the end of the deck calibration program in the app. Please make sure we are OK with this language before merging.

[sfoster1]

Python side looks good (haven't tested though) but we should change the FE copy a bit

[sfoster1]

Maybe cut this para as discussed yesterday? I think just the final para ("In order to effectively use..." is enough of a warning.

[Laura-Danielle]

Ah ok, I misunderstood. Will remove it.

[sfoster1]

LGTM pending QA and UX review

[codecov]

Codecov Report

❗ No coverage uploaded for pull request base (edge@0e1b15b). Click here to learn what that means.
The diff coverage is n/a.

@@           Coverage Diff           @@
##             edge    #6208   +/-   ##
=======================================
  Coverage        ?   79.58%           
=======================================
  Files           ?      199           
  Lines           ?    18420           
  Branches        ?        0           
=======================================
  Hits            ?    14660           
  Misses          ?     3760           
  Partials        ?        0           

Impacted Files	Coverage Δ
opentrons/protocol_api/geometry.py	93.60% <0.00%> (ø)
opentrons/calibration_storage/types.py	100.00% <0.00%> (ø)
opentrons/config/feature_flags.py	94.11% <0.00%> (ø)
..._server/service/legacy/routers/deck_calibration.py	95.45% <0.00%> (ø)
...ver/service/session/session_types/check_session.py	95.34% <0.00%> (ø)
...-server/robot_server/service/legacy/models/logs.py	100.00% <0.00%> (ø)
...r/robot_server/service/legacy/models/networking.py	100.00% <0.00%> (ø)
opentrons/legacy_api/robot/mover.py	88.63% <0.00%> (ø)
opentrons/commands/types.py	100.00% <0.00%> (ø)
update-server/otupdate/buildroot/config.py	92.50% <0.00%> (ø)
... and 189 more

---

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 0e1b15b...4631945. Read the comment docs.

[nusrat813]

Tested - working as intended

[SyntaxColoring]

@Laura-Danielle

by resetting instrument offset for all pipettes any time a new deck calibration begins

If you start and then abort a deck calibration, is the pipette calibration still cleared? (If so, I don't think it would be a big problem—just looking out for surprises.)