eval: add eval_results.json, model-index, fix evaluate.py NaN→null

eval: add eval_results.json, model-index, fix evaluate.py NaN→null

eval: add eval_results.json, model-index, fix evaluate.py NaN→null

eval: restore full results + add mamba-main

eval: restore full results + add mamba-main

eval: add mamba-main benchmark results

eval: add mamba-main benchmark results

eval: add self-contained evaluate.py with HF push support

eval: fix eval_results.json schema for ModelDashboard

docs: clean benchmark section + proper model-index (mean metrics, no per-dataset wall)

eval: add visual comparison spot.png

eval: add visual comparison naip.png

eval: add visual comparison spain_crops.png

eval: add visual comparison spain_urban.png

eval: add eval_results.json for HF Dashboard

eval: add model-index eval results + benchmark section to README

eval: add benchmark bar chart

eval: add benchmark results CSV (opensr-test, 3 variants)

Add evaluate.py and variant field in config.json

Add WEO-SAS model card

Update sen2sr_pt.py for multi-variant support

Update config.json for multi-variant support

Add SEN2SRLite RGBN x4 with WEO standard interface

initial commit