Explore - Support week-over-week performance comparison with weekly time periods and intervals

### Problem Statement

Currently, performance metrics are limited to 1-day time periods and intervals, which
creates too much noise for meaningful trend analysis and reporting. Daily data is
affected by holidays, releases, reverts, and other one-off events, making it difficult
to identify genuine performance improvements over longer timeframes. This prevents teams
from accurately measuring and reporting on performance changes for OKRs and quarterly
reviews, where consistent week-over-week or quarter-over-quarter comparisons are needed
to demonstrate real progress (e.g., "we reduced LCP by X% compared to last quarter").

### Solution Brainstorm

_No response_

### Product Area

Explore

No files, tests, or entry points are named. Start by tracing the Explore performance-metrics implementation and defining how weekly time periods and intervals should be represented for comparisons. Done should include meaningful week-over-week comparisons, including metrics such as LCP, with behavior suitable for trend analysis and reporting.

Written by the indexing model from the issue text.