Update docs: challenges.md (training bias + scene text approach), vam/multimodal/trainers docs sync

Revert camera prompt changes; add NaN guard and CUDA sync; add engineering challenges doc

refactor: reorganize dataset into lm/vlm/vam subdirectories

docs: update README and docs for model_type rename, RoPE guard fix, config paths

reorganize configs into subdirs lm/vlm/vam, remove runs/, add interview docs for pretrain & full_sft

docs: move tokenizer.md into interview/

refactor(docs): restructure docs with modular organization

Replace generic README.md doc names with descriptive index/overview names

Add modular development docs under docs/ for learning and interviews