v2.1.1 - Community setup and repository structure update

This release improves the repository structure and adds community files that help users discover, cite, report issues, and contribute to the project.

Added
Added root-level README.md so the GitHub repository homepage shows project usage and onboarding details directly.
Added MIT LICENSE.
Added CITATION.cff so GitHub can generate citation metadata.
Added CONTRIBUTING.md with bug report and pull request guidance.
Added CODE_OF_CONDUCT.md.
Added SECURITY.md with API key and privacy guidance.
Added GitHub Issue Forms for bug reports, feature requests, and questions.
Added a pull request template.
Added synthetic PubMed .nbib and Scopus .csv sample files under sample_data/.
Added docs/OUTREACH.md with short announcement text and recommended GitHub topics.
Changed
Moved the main application files to the repository root for easier installation and execution after cloning.
Updated README content with badges, quick start steps, PubMed enrichment explanation, citation instructions, contribution links, and roadmap items.
v2.1.0 - Encoding Fixes and PubMed Reference Enrichment
This release fixes issues found after the initial public upload and adds a major PubMed reference completion workflow.

Fixed
Fixed mojibake/garbled Chinese text across code comments, GUI labels, log messages, dialog messages, and AI prompts.
Replaced broken or unclear interface strings with readable Chinese text.
Made the OpenAI dependency optional at runtime so the GUI can still start when the openai package is not installed.
Improved GUI thread handling for log updates and error dialogs.
Added
Added PubMed DOI-based lookup through NCBI E-utilities.
Added DOI to PMID search for PubMed records.
Added PubMed reference retrieval using linked PubMed reference records.
Added fallback parsing for reference lists returned inside PubMed XML when available.
Added conversion of PubMed reference metadata into WOS-compatible CR cited-reference lines.
Added automatic NR reference count generation.
Added NCBI Email and NCBI API Key fields in the GUI.
Added request throttling and retry handling for NCBI calls.
Added WOS export header lines: FN Clarivate Analytics Web of Science and VR 1.0.
Added a more stable WOS field output order.
Improved
Improved PubMed .nbib parsing:
extracts PMID;
extracts DOI from LID, AID, or DOI-like text;
preserves short author names and full author names;
extracts journal title and journal abbreviation;
extracts abstract, publication year, publication date, volume, issue, pages, ISSN, keywords, MeSH terms, language, and publication type.
Improved Scopus .csv parsing and reference preservation.
Improved DOI normalization and text matching.
Improved deduplication:
deduplicates by DOI;
falls back to title + year + first author;
merges useful fields from duplicate records instead of simply discarding them;
preserves and merges CR reference lists where possible.
Improved WOS export formatting for list fields and multiline fields.
Updated the GUI layout to include separate PubMed reference enrichment and OpenAI extraction sections.
Notes
PubMed .nbib exports do not normally contain full cited-reference data. The new enrichment workflow uses DOI/PMID to query PubMed and adds the linked references that PubMed can provide. Reference completeness still depends on PubMed indexing and link availability.