Scaling translation operations in wagtail-localize

This project focuses on improving the performance of core processes in wagtail-localize, particularly in the construction of the page index and in translation workflows that operate on large page trees. Current analysis of the codebase reveals patterns that lead to repeated database queries and sequential processing, which can negatively impact performance as the number of pages and locales grow. These issues become especially noticeable in large, content-heavy sites, where administrative actions such as translating or updating page subtrees can become slow and less reliable. The project will begin by measuring query counts and execution time for key operations, establishing a baseline for comparison. Based on these measurements, the page index construction will be refactored to reduce database queries by fetching data more efficiently and processing it in memory. Translation workflows will be analyzed to identify opportunities to reduce sequential processing and improve performance when processing large page trees. Additionally, the current mechanism for detecting content changes will be reviewed to reduce unnecessary work when no updates are required. The main deliverables include characterization tests and benchmarks, a refactored page index implementation with reduced query counts, improvements to translation workflows, and an evaluation and potential optimization of content change detection. All improvements will be validated against the established baseline to ensure measurable performance gains.