feat(food): improve food search with diacritic-insensitive and composed multi-word matching

[marcuscastelo]

This PR enhances the food search functionality to address issues where searches for multi-word or diacritic-containing food names (e.g., "Tapioca doce") returned no results. The main changes are:

• Food search is now diacritic-insensitive, ensuring queries match regardless of accents or special characters.
• Implements a composed search: first returns exact matches for the full query, then includes partial matches for each word in the query, removing duplicates.
• Adds a removeDiacritics utility and corresponding tests to support normalization.
• Updates the repository logic to use the new composed and normalized search.
• Refactors and tests the search logic for robustness.

No breaking changes.
closes #745

[vercel]

The latest updates on your projects. Learn more about Vercel for Git ↗︎

Name	Status	Preview	Comments	Updated (UTC)
marucs-diet	✅ Ready (Inspect)	Visit Preview	💬 Add feedback	Jun 12, 2025 9:52am

[Copilot]

Pull Request Overview

Enhances food search to handle diacritics and multi-word queries by normalizing input and composing exact and partial matches.

• Adds a removeDiacritics utility with tests for normalization.
• Refactors fetchFoodsByName and internalCachedSearchFoods to apply normalization and multi-word logic.
• Updates prompt file with milestone complexity labeling instructions.

Reviewed Changes

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

File	Description
src/shared/utils/removeDiacritics.ts	Implements diacritic removal utility.
src/shared/utils/removeDiacritics.test.ts	Adds unit tests for the diacritic removal function.
src/modules/diet/food/infrastructure/supabaseFoodRepository.ts	Normalizes search terms and composes multi-word searches.
.github/prompts/milestone-complexity-labeler.prompt.md	Introduces milestone complexity labeling prompt.

Comments suppressed due to low confidence (2)

src/modules/diet/food/infrastructure/supabaseFoodRepository.ts:201

• Filtering with Array.some inside filter leads to O(n²) behavior. You can use a Set of exact match IDs (e.g., const exactIds = new Set(exactMatches.map(f => f.id))) for O(1) lookups and better performance.

partialMatches = partialResults.flat().filter((food) => !exactMatches.some((f) => f.id === food.id))

src/shared/utils/removeDiacritics.test.ts:6

• Consider adding tests for edge cases such as strings with multiple diacritics in different scripts or composed characters (e.g., 'Ångström') to ensure full coverage.

it('removes accents from common pt-BR words', () => {