E2E test: workaround bad kernels (uv managed)

[testlabauto]

Need to skip interpreters that tests must not use.

@:positron-notebooks
@:package-pane

[github-actions]

E2E Tests 🚀
This PR will run tests tagged with: @:critical @:positron-notebooks @:package-pane

^{readme  valid tags}

[github-actions]

PETE's assessment 🧪

Verdict: 🟡 Not applicable -- this PR only edits the shared e2e fixture settings.json to exclude ~/.local interpreters/kernels from test runs; it introduces no product behavior change to assert against.

What changed

• Adds jupyter.kernels.excludePythonEnvironments and python.interpreters.exclude (both ["~/.local"]) to test/e2e/fixtures/settings.json so e2e runs skip interpreters that tests must not pick up. This is test-harness configuration, not a source change.

Tests in this PR

• Unit (Vitest/Mocha) ✅ (not applicable -- no source behavior changed)
• Extension host ✅ (not applicable)
• E2E (Playwright) ✅ (this is an e2e fixture change; both keys are real -- defined in extensions/positron-python/package.json)

Existing coverage

N/A -- a fixture/config edit has no behavior of its own to cover. The settings file feeds existing e2e suites (loaded via test/e2e/fixtures/test-setup/*); those suites are the consumers and remain the coverage for any interpreter-selection behavior.

Suggested additions

None.

Suggested tags (optional)

The PR body already lists @:positron-notebooks and @:package-pane, which is appropriate for a fixture change affecting kernel/interpreter selection in those areas. No additional feature tags warranted -- the change is broad infra rather than scoped to one feature.

---

PETE (Positron Extreme Test Experiment) - LLM-based test-coverage advisor, in pilot. Triggers on PR open and on /recheck-tests comments. Wrong verdict? Comment /recheck-tests (or /rePETE) on this PR to re-run. Please share feedback on how PETE performed here.

[isabelizimm]

will windows respect the ~? I am not sure how it works for settings 👾

[jonvanausdeln]

I don't know if it will expand correctly, but for this use it won't be an issue. I just checked locally and it doesn't break anything.

[isabelizimm]

approving to not be the blocker here, but not sure if it needs to be a different path for windows e2e tests