can't render content-based recommendation on frontend.

[tindo2003]

I modified builder/lda_model_calculator.py as follows:

x_id = str(docs[x].movie_id)
y_id = str(docs[y].movie_id)

was updated to:

x_id = str(docs[x].imdb_id)
y_id = str(docs[y].imdb_id)

Reason for the change:
In recs/content_based_recommender.py, the function recommend_items_by_ratings returns data in the following structure:

{
    [
        <target_id>,
        {
            "prediction": "9.5500",
            "sim_items": [
                <source_id>, <source_id>, ...
            ]
        }
    ]
}

The frontend template /templates/moviegeek/index.html expects <target_id> to be an IMDb ID so it can use the get_url() function in base.html to fetch information about the recommended movie.

Previously, the old code could not fetch the correct movie information because the target and source fields in the lda_similarity table of the database stored TMDb IDs instead of IMDb IDs. By changing movie_id to imdb_id, the recommendation system now aligns with the frontend's requirements and ensures proper functionality.

The other cosmestic changes are due to my linter (my bad)