Fix FIXME and avoid override zips with the same original name

[lfoppiano]

• Fixed the filenames convention as {SCOPE}-REPACKAGED-{ORIGINAL_NAME}.warc.gz
• Fixed issue with overriding zip files (reported in confusing names #13
• Updated instructions for massaging the manifest.csv + download
  • Added a new command warc2zip manifest.csv --fetch

[lfoppiano]

The hex appended will match the directory.

$ warc2zip s3://commoncrawl/crawl-data/CC-MAIN-2026-34/segments/1786091384908.68/crawldiagnostics/CC-MAIN-20260807101845-20260807131845-00000.warc.gz
Reading WARC: 27.0MB [00:04, 5.54MB/s]
Warning: 150 orphan group(s) without a response record, skipped
Created CC-MAIN-20260807101845-20260807131845-00000_44ca.zip: 3639 responses, 3789 requests, 3789 metadata records

$ warc2zip s3://commoncrawl/crawl-data/CC-MAIN-2026-34/segments/1786091384908.68/robotstxt/CC-MAIN-20260807101845-20260807131845-00000.warc.gz      
Reading WARC: 1.49MB [00:00, 1.61MB/s]
Created CC-MAIN-20260807101845-20260807131845-00000_2a58.zip: 833 responses, 833 requests, 0 metadata records

The directory name will be slighly different because it contains also the timestamp, so it will partially match the original zip file:

$ zipinfo -1 CC-MAIN-20260807101845-20260807131845-00000_2a58.zip | cut -d/ -f1 | sort -u 
CC-MAIN-20260807101845-20260807131845-00000_20260904T074146_2a58
$ zipinfo -1 CC-MAIN-20260807101845-20260807131845-00000_44ca.zip | cut -d/ -f1 | sort -u  
CC-MAIN-20260807101845-20260807131845-00000_20260904T074137_44ca