Refactor Keithley 2600 driver to use ParameterWithSetpoints for fast sweep

[trlemon]

Background

The doFastSweep method relies on "old" technology from qcodes_loop, and it also returns a Qcodes dataset which may not be desirable for all applications. Instead, it would be better to use ParameterWithSetpoints and return a numpy array.

Summary of Changes

• Remove dependence on qcodes_loop
• Use ParameterWithSetpoints with additional parameters for configuration.
• Return fast sweep data as numpy array instead of qcodes dataset.
• Add functionality to perform both 1D (single channel) and 2D (two channels) buffered sweeps.

[codecov]

Codecov Report

❌ Patch coverage is 84.57447% with 29 lines in your changes missing coverage. Please review.
✅ Project coverage is 70.30%. Comparing base (b0a7b55) to head (6c5f43e).
⚠️ Report is 83 commits behind head on main.

Files with missing lines	Patch %	Lines
...odes/instrument_drivers/Keithley/_Keithley_2600.py	84.57%	29 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #7851      +/-   ##
==========================================
+ Coverage   60.56%   70.30%   +9.73%     
==========================================
  Files         333      333              
  Lines       32245    32401     +156     
==========================================
+ Hits        19529    22778    +3249     
+ Misses      12716     9623    -3093     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[astafan8]

looks promising!

[astafan8]

is it ready for the final review, can it exit the draft mode? :)

[jenshnielsen]

@trlemon This needs a rebase / merge probably because of the 4 wire pr that was merged

[Copilot]

Pull request overview

Refactors the Keithley 2600 fast sweep implementation away from qcodes_loop by introducing a ParameterWithSetpoints-based API (setup_fastsweep + fastsweep) that returns NumPy arrays and supports both 1D and 2D buffered sweeps.

Changes:

• Added setup_fastsweep configuration method and rewrote fast sweep execution to use on-instrument Lua scripts via ParameterWithSetpoints.
• Added 1D/2D fast sweep configuration + execution tests (including channel detection and reconfiguration regression coverage).
• Added a changelog newsfragment describing the driver change (and marking it as breaking).

Reviewed changes

Copilot reviewed 3 out of 4 changed files in this pull request and generated 4 comments.

File	Description
tests/drivers/test_keithley_26xx.py	Adds tests for 1D/2D fast sweep configuration, execution shaping, channel inference, and reconfiguration regression.
src/qcodes/instrument_drivers/Keithley/_Keithley_2600.py	Introduces _FastSweepConfig, setup_fastsweep, and rewrites fastsweep as a ParameterWithSetpoints using Lua scripts.
docs/changes/newsfragments/7851.improved_driver	Documents the breaking fast sweep API change in the changelog.

[Copilot]

Pull request overview

Copilot reviewed 3 out of 4 changed files in this pull request and generated 5 comments.

[Copilot]

_LinSweepLike.param is typed as ParameterBase, but setup_fastsweep immediately casts it to Parameter and passes it to infer_channel, whose signature requires a Parameter (see qcodes/extensions/infer.py). This mismatch means the protocol currently claims to accept sweep objects whose param is not a Parameter, but the implementation will fail at runtime / type-check time. Please align the protocol with actual requirements (e.g., require param: Parameter) and remove the unsafe cast, or adjust the channel inference to support ParameterBase inputs.

[jenshnielsen]

@trlemon Could you have a look at the last copilot review comments. Other than that this should be good to merge