feat(search.ts): improve search ranking by incorporating popularity score

[GruperTal]

Description

Improves search result ranking by incorporating TMDB popularity scores into the sorting algorithm. Previously, search results were returned in TMDB's default order which prioritizes text matching over popularity, causing obscure titles to appear above well-known ones.

Problem
When searching for partial titles (e.g., "the hunger"), less popular movies with closer text matches would appear before highly popular titles like "The Hunger Games" franchise.

Solution
Implemented a scoring system that combines:
Popularity (primary factor) - TMDB popularity score weighted heavily
Text relevance (secondary factor) - small boosts for titles starting with or exactly matching the query
This ensures popular titles rank appropriately while still giving slight preference to closer text matches when popularity is similar.

Changes
Added calculateSearchScore() function that computes a weighted score
Added sortResultsByRelevanceAndPopularity() to sort results by combined score
Applied sorting to all search results including special search providers (year:, tmdb:, etc.)

Testing
Tested with queries like "the hunger" - The Hunger Games movies now appear before obscure titles like "The Hunger".

To-Dos

• Successful build yarn build

[fallenbagel]

This is a good QOL feature, imho. Please redirect your PR towards https://github.com/seerr-team/seerr as this repo is no longer maintained (we will archive this repo soon).

@GruperTal