SEO Optimization: enhance keyword tags (bhojpuri, bho, bhoj, devanagari, indic) and title search indexing

Refine tags: keep only specific descriptive tags

Improve discoverability: add bhojpuri, devanagari, nlp tags and HF metadata fields

Fix: author name to Satyam Kumar, fix CITATION.cff repository URL

Fix: add HF dataset viewer config, correct username in README

Initial release: BhojpuriCorpus v1.0.0 — 386032 documents, 24.97M estimated tokens