data(cpu): harvest AMD Eng Sample OPNs

[Seungpyo1007]

What & why

Slice 1 leftover LGA1151 ES rows were already in the catalog, so this fire fell through to AMD news OPNs explicitly labeled Engineering Sample / Eng Sample.

Added (6 ES):

• 100-000000560-40_Y — Rembrandt 8C/16T laptop (Tom's Hardware)
• 100-000001290-11_N — Granite Ridge 8C/16T (Wccftech / Einstein@Home)
• 100-000000884-21_N — Storm Peak 96C/192T Threadripper (Wccftech)
• 100-000000884-20_Y — Storm Peak 96C/192T sibling
• 100-000000666-20_Y — Raphael 8C/16T (PC Gamer)
• 100-000000719-52_Y — Granite Ridge 16C/32T 3D V-Cache (VideoCardz)

Skipped: leftover LGA1151 complete-cpu-list/coffee-mod (already harvested); Coffee Lake QS QNMM/QNMK/QQPL/QQPK/QQPP; mixed LGA2011-3 tables; Fire Range 100-000001028-42_Y (threads not published).

QS count in this PR: 0.

Catalog: 126 -> 132 ES records.

Refs #1

Source

• https://www.tomshardware.com/news/amds-first-ryzen-cpu-with-alleged-ddr5-support-emerges
• https://wccftech.com/amd-zen-5-granite-ridge-ryzen-desktop-cpu-8-cores-16-threads-pictured-leak/
• https://wccftech.com/amd-ryzen-threadripper-7000-storm-peak-cpu-96-zen-4-cores-480-mb-cache-spotted/
• https://www.pcgamer.com/52ghz-amd-zen-4-test-cpu-pops-up-in-database-and-then-vanishes/
• https://videocardz.com/newz/amd-16-core-ryzen-9000hx-fire-range-and-ryzen-9-9950x3d-cpus-spotted-in-geekbench-ai-test

Checklist

• python -m app.validate passes
• every catalog file classifies as es
• QS count in this PR: 0