Used Japanese tokenization in queries for accurate word searching in …

[H3xano]

Overview

This PR addresses the issue where searching for compound Japanese terms like "標準製品" did not return the expected matches in Lunr.js. The solution involves:

Japanese Morphological Segmentation: Utilizes the lunr.ja plugin to ensure consistent tokenization of Japanese text during both indexing and querying.

Wildcard Query Enhancement: Appends trailing wildcards (*) to each token in the search query to enable partial and prefix matching.

Error Handling: Ensures that queries are constructed using Lunr's query builder API to prevent parsing errors and maintain consistency.

Changes Made

• Reintroduced the lunr.ja plugin for accurate Japanese tokenization.
• Modified the Search component to tokenize queries using the same pipeline.
• Utilized Lunr's query builder API to construct searches programmatically, ensuring error-free query execution.

[strogonoff]

There was no need to convert from ESM to CJS… And it makes checking the changes more difficult ;) @H3xano