Collecting data from Greek marketplace

I want to collect complete product data from a specific e-commerce site—around 50,000-60,000 SKUs divided into each category, with at least 50 items per subcategory, and broken down by gender (male/female) and manufacturer. The script will output the basics in a csv file: product name, full category tree, short and long descriptions, technical specifications, color, size, product manufacturer, price, store name, store URL, and most importantly, one or more image URLs per item.

What I still need is a clean, reliable Python 3 script that will contain:
All of the above data

• Fetch each image URL directly to my server, storing files in a logical folder hierarchy that reflects gender → main category → subcategory.

• Keeps the file names unique (product code or other field that already exists).

• Records successes and failures, with an easy option to resume if the process is interrupted. • Efficiently handle at least 60,000 products—identity downloads or asynchronous downloads (requests, aiohttp, or similar) are welcome, as long as they remain lightweight.

Validate each download (status 200, file not empty) and retry a configurable number of times before marking it as failed.

• Generate a final report (CSV or TXT) of skipped or failed data + image, so I can review them later.

Feel free to suggest optimal libraries or modifications, but keep dependencies minimal and typical (requests, asyncio, aiohttp, urllib3, pandas, etc.). The script will run on a Linux VPS. Assume Python 3.10 is available.

Deliverables
1. The complete, well-commented Python script. The full data results. A separate CSV file with all categories, subcategory URLs, and the total number of products each category and subcategory has in this marketplace.
2. A short README or step-by-step instructions on how to configure paths, credentials (if any), and run the script via CLI or Cron to download the image.

3. Optional: a short test run on a subset of 500 products to demonstrate speed and reliability.