CI/CD: Duplicate issue detector and labeller as "duplicate"

[aniket866]

Is your feature request related to a problem?

• Yes, it is related to a problem

Describe the feature you'd like

Problem: Contributors must manually search through many existing issues before creating a new one, which is time-consuming and often leads to duplicate reports.

How: Add automated duplicate detection using a GitHub bot (semantic matching) and a lightweight GitHub Action to flag similar existing issues on creation.

Impact: it will mark that issue duplicate so that maintainer don't have to waste too much time

Notes: Automation will only suggest possible duplicates (no auto-closing, or auto-closing depends on choice) .

Record

• I agree to follow this project's Code of Conduct
• I want to work on implementing this feature

[ML-dev-crypto]

Hi maintainers, I'd like to take on this issue (assign to @ML-dev-crypto) and implement an automated duplicate detector that labels new issues as "duplicate" and comments with links to similar reports.

Proposed solution:

• Approach: Combine a lightweight GitHub Action (triggered on issues.opened and issues.edited) with a semantic similarity check across existing open issues.
• Detection:
  • Use a small Python script with sentence-transformers (or a fast embedding model) to compute embeddings for the new issue’s title/body and compare against embeddings of recent/open issues via cosine similarity.
  • Fallback to fuzzy keyword matching if embedding setup fails (to keep action resilient).
• Action behavior:
  • If similarity exceeds a configurable threshold (e.g., 0.85), automatically:
    • Add the duplicate label to the new issue (configurable label name).
    • Post a comment listing the top N similar issues with links and similarity scores.
  • No auto-closing by default; maintainers can manually close if appropriate. Auto-close can be toggled via workflow inputs.
• Configuration & safety:
  • Configurable parameters: similarity threshold, max issues to compare, top N suggestions, label name, auto-close toggle.
  • Respect rate limits; cache embeddings for recent issues to minimize API calls.
  • Document setup, thresholds, and maintenance steps for future adjustments.