fix: tighten role inference and allow cRoles editing

[michaelmwu]

Description

• Tightened resume-role prompt guidance so designer is inferred only from explicit design-role titles, avoiding false positives from generic design wording.
• Added regression test coverage for action-language design text.
• Added Discord UX to edit suggested cRoles with a new modal, button, and normalization handling in the resume confirmation flow.
• Extended tests for cRoles modal visibility, prefill, submit, clear, and callback behavior.

Related Issue

How Has This Been Tested?

• Pre-commit checks (ruff + mypy) pass during commit.
• Added unit tests in tests/unit/test_crm.py and tests/unit/test_resume_extractor.py.

Summary by CodeRabbit

• New Features

  • Users can now edit proposed role changes through an interactive modal during the resume update confirmation process.

• Bug Fixes

  • Refined role detection to only recognize "designer" when explicitly stated in job titles or headings, preventing false detection from design-related action verbs.

[coderabbitai]

Caution

Review failed

Pull request was closed or merged during review

📝 Walkthrough

Walkthrough

This PR adds UI components for editing resume roles within the CRM discord bot and constrains designer role inference to explicit design-role signals rather than generic action verbs. It includes a new modal and button for role editing, role normalization utilities, and corresponding test coverage for both the UI additions and inference constraint.

Changes

Cohort / File(s)	Summary
Discord Bot Roles UI apps/discord_bot/src/five08/discord_bot/cogs/crm.py	Added ResumeEditRolesModal and ResumeEditRolesButton classes; integrated into ResumeUpdateConfirmationView with conditional visibility when cRoles are proposed; added _normalize_roles_value() static method for role normalization.
Resume Extractor Role Inference packages/shared/src/five08/resume_extractor.py	Narrowed designer role inference constraint to require explicit design-role signals (e.g., Product Designer, UI Designer) rather than inferring from generic "designed" action verbs.
CRM UI Component Tests tests/unit/test_crm.py	Added test coverage for ResumeEditRolesButton and ResumeEditRolesModal, verifying button visibility, modal prepopulation, submission behavior, and removal of roles from proposed updates.
Resume Extractor Inference Tests tests/unit/test_resume_extractor.py	Added test_infer_roles_does_not_treat_design_verbs_as_designer() to verify design verbs do not incorrectly map to designer role.

Possibly related PRs

• fix: improve resume location parsing and editing #180: Modifies the same ResumeUpdateConfirmationView class, adding similar Edit UI components and normalization helpers for location fields.
• Improve resume CRM inference and Espo client integration #168: Extensively modifies role inference in the same resume_extractor.py file, directly overlapping with this PR's designer inference constraints.
• Add skills edit modal and skills delta preview #166: Adds comparable UI components and normalization helpers to ResumeUpdateConfirmationView, following the same pattern for skills fields.

Estimated code review effort

🎯 3 (Moderate) | ⏱️ ~20 minutes

---

🐰 Roles get their editing buttons bright,
Designers now need clearer light,
No more verbs to claim the crown,
Just explicit signals matter now,
Tests confirm it all works right! 🌟

🚥 Pre-merge checks | ✅ 2 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 60.00% which is insufficient. The required threshold is 80.00%.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (2 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title accurately summarizes the two main changes: tightening role inference logic and adding UI capability to edit cRoles in the confirmation flow.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

• 📝 Generate docstrings (stacked PR)
• 📝 Generate docstrings (commit on current branch)

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Post copyable unit tests in a comment
• Commit unit tests in branch michaelmwu/resume-role-fix

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[Copilot]

Pull request overview

This PR tightens resume role inference around “designer” to reduce false positives from action-language “design” wording, and adds a Discord confirmation-flow UI to edit proposed CRM roles (cRoles) before applying updates.

Changes:

• Updated the shared resume-extractor prompt guidance to infer designer only from explicit design-role signals (not generic “design/designed” language).
• Added regression tests ensuring action-language “design” wording does not infer the designer role.
• Added Discord UI components (button + modal) to allow editing proposed cRoles, plus unit tests covering visibility, prefill, submit, and clear behaviors.

Reviewed changes

Copilot reviewed 4 out of 4 changed files in this pull request and generated 1 comment.

File	Description
tests/unit/test_resume_extractor.py	Adds regression test preventing “design” action-language from inferring designer.
tests/unit/test_crm.py	Adds unit coverage for cRoles edit button/modal behavior in the Discord confirmation view.
packages/shared/src/five08/resume_extractor.py	Tightens LLM prompt instructions for designer inference.
apps/discord_bot/src/five08/discord_bot/cogs/crm.py	Adds ResumeEditRolesModal, ResumeEditRolesButton, view wiring, and role normalization helper.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

normalize_roles already supports splitting a string on commas/newlines/semicolons. Here the modal pre-splits into a list (and only replaces commas), so inputs like developer;marketing (or marketing; designer) will be treated as a single role token and may normalize incorrectly. Consider passing the raw text string directly to normalize_roles(...) (or also splitting on ;) and letting it handle delimiter parsing/deduping.

Suggested change

	cleaned = [
	line.strip() for line in raw.replace(",", "\n").splitlines() if line.strip()
	]
	normalized_roles = normalize_roles(cleaned)
	normalized_roles = normalize_roles(raw)