feat(robot): allow conditional no-status-rc

[axisrow]

Summary

Allow no-status-rc to accept the existing Condition form, enabling branch-dependent status-code behavior without duplicating profiles:

default-profiles = ["ci"]

[profiles.ci]
enabled = { if = "environ.get('CI') == 'true'" }
no-status-rc = { if = "environ.get('CI_COMMIT_REF_NAME') == 'main'" }

A failing Robot run now returns 0 on main and preserves the failure status on feature branches. Existing true, false, on, off, and default behavior remains unchanged.

Fixes #482.

Design choice

I prototyped both designs raised in the issue against the same 12 behavioral cases:

• Reuse Condition { if = ... }.
• Add a public BoolExpression { expr = ... } type.

Both produced the required boolean flag behavior. I kept Condition because it matches the requested syntax, reuses the existing safe evaluation path, and avoids another public model and schema definition. The conditional form resolves to true or false; static default remains available through the existing literal form.

Because Condition is now shared by profile fields and a flag field, the schema generator also avoids attaching a context-specific documentation link to its shared if property.

Type of change

• Bug fix
• New feature
• Refactor / internal change
• Documentation
• Build / CI / tooling
• Breaking change

Checklist

• I have read the Contribution Guide and the AI and Automated Contribution Policy.
• The change is focused on a single concern (no unrelated refactors or formatting noise).
• Tests for the change have been added or updated, and hatch run test:test passes locally (RF matrix against the default Python) — or N/A for documentation-only / non-code changes. See Running Tests for faster iteration options and the full matrix.
• hatch run lint:all passes (also enforced by the pre-commit hooks if installed) — or N/A for documentation-only / non-code changes.
• Documentation has been updated where relevant (user docs, code comments, README).
• Generated files (if any) were regenerated with the documented script, not edited by hand.
• Commits follow Conventional Commits and are cryptographically signed (git commit -S, GPG/SSH).

AI / tooling disclosure

• No AI/automation was used beyond ordinary editor autocomplete.
• AI/automation was used; tool(s): OpenAI Codex through Agent Orchestrator
  • I reviewed the generated output manually, understand it, and verified it locally.

AI/tooling disclosure: This contribution was prepared with assistance from OpenAI Codex through Agent Orchestrator. I manually reviewed the diff, understand the implementation, verified it locally, and confirm that I have the right to submit it under the project license.

Additional notes

Acceptance proof

The end-to-end test uses the configuration shown above, selects the single ci profile through default-profiles, and runs a deliberately failing Robot suite through a real RobotCode subprocess without -p:

• CI=true, CI_COMMIT_REF_NAME=main: --nostatusrc, exit code 0.
• CI=true, CI_COMMIT_REF_NAME=feature/example: --statusrc, exit code 1.

The same test fails on unmodified main while parsing profiles.ci.no_status_rc because the conditional table is not accepted there.

Verification

• hatch run test:test: passed across rf50, rf60, rf61, and rf70 through rf74; the final rf74 run reported 3933 passed, 22 skipped, and 2 xfailed.
• Focused matrix: 28 tests passed in each of the eight Robot Framework environments.
• Ruff check and formatting stages: passed; 413 files already formatted.
• JSON Schema Draft 7 validation: passed.
• Generator idempotence: repeated model and schema generation produced identical hashes.
• git diff --check: passed.

A fresh hatch run lint:all reaches one mypy 2.2.0 comparison-overlap error in the untouched packages/plugin/src/robotcode/plugin/__init__.py:352. The same error reproduces on unmodified main; Ruff and all feature-related tests pass. The lint checklist remains intentionally unchecked.

Generated artifacts

The model and schema were regenerated with:

hatch run test.rf74:python scripts/generate_rf_options.py
hatch run test.rf74:ruff format packages/robot/src/robotcode/robot/config/model.py
hatch run test.rf74:python scripts/create_robot_toml_json_schema.py

The updated no-status-rc documentation matches the generated model output. Generator compatibility adjustments preserve existing field aliases, the existing max-error-lines = "NONE" example, and generated-block spacing so regeneration does not introduce unrelated output drift.