CLN: Use ruff to improve code quality

[bashtage]

No description provided.

To address the issue, replace the ... in the method body of keys in SupportsKeysAndGetItem (line 26) with a pass statement. This fixes the "statement has no effect" warning and follows Python convention for abstract stub methods. Only change the method body; do not touch the method signature or docstring. No other changes or imports are necessary. Since this is a protocol intended for type-checking/interface definition, a pass statement is the clearest and standard way to indicate that the method has no runtime body.

---

The best way to fix the problem is to remove the unused import statement import matplotlib as mpl # noqa: F401 on line 16 in linearmodels/tests/test_examples.py. This can be done by simply deleting line 16, taking care not to touch any surrounding lines or related imports for libraries that are indeed used (jupyter_client, ExecutePreprocessor, nbformat, etc.). This change does not affect any existing functionality, as mpl is not referenced elsewhere.

To resolve this issue, simply remove the unused import statement on line 19 (import seaborn as sns # noqa: F401). No other changes are needed since seaborn is not used in the code. Check that you only remove the specific line and do not affect any other logic or imports in the file. There’s no need to add, replace, or update any other code elsewhere.

[codecov]

Codecov Report

❌ Patch coverage is 98.43750% with 9 lines in your changes missing coverage. Please review.
✅ Project coverage is 99.57%. Comparing base (287b79c) to head (b8e840b).
⚠️ Report is 5 commits behind head on main.

Files with missing lines	Patch %	Lines
linearmodels/panel/data.py	85.71%	0 Missing and 3 partials ⚠️
linearmodels/shared/hypotheses.py	33.33%	2 Missing ⚠️
linearmodels/shared/utility.py	50.00%	0 Missing and 2 partials ⚠️
linearmodels/system/model.py	92.00%	1 Missing and 1 partial ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #661      +/-   ##
==========================================
- Coverage   99.57%   99.57%   -0.01%     
==========================================
  Files          99       99              
  Lines       17333    17278      -55     
  Branches     1444     1419      -25     
==========================================
- Hits        17260    17205      -55     
  Misses         27       27              
  Partials       46       46              

Flag	Coverage Δ
adder	99.56% <98.43%> (-0.01%)	⬇️
subtractor	99.56% <98.43%> (-0.01%)	⬇️

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

☔ 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.

To fix the cyclic import, we should avoid importing from linearmodels.panel.model inside linearmodels.panel.results. Since only _deferred_f is needed, and only within the f_statistic_robust method, we can move _deferred_f out of linearmodels.panel.model and into a new module (e.g., linearmodels.panel._shared_fstat.py). Then, both results.py and model.py can import it without creating a cycle. In this fix, since we can only edit the contents of results.py, the best approach is to move the local import to the top of the file (module-level import), but change the import path to a completely new module—assuming _deferred_f has been moved there. We will update the local import in f_statistic_robust to a top-level import from linearmodels.panel._shared_fstat, and replace

from linearmodels.panel.model import _deferred_f  # noqa: PLC0415

with

from linearmodels.panel._shared_fstat import _deferred_f

and add this import at the top of the file to fully break the cycle.
All usages of _deferred_f will continue working as before, but now without the cyclic dependency.

---

The fix is to remove the from linearmodels.compat.statsmodels import Summary statement from the file, as the module is already imported with import linearmodels.compat.statsmodels. In any place where the code uses Summary directly (i.e., as Summary(...)), we should instead use linearmodels.compat.statsmodels.Summary(...). The imports at the top of the file should be edited to delete the from ... import Summary line. Search through the file for usages of Summary and update them to fully reference linearmodels.compat.statsmodels.Summary. No additional imports are needed.