feat: add machine-readable JSON output mode

[KGFCH2]

@Rakshat28

Addresses #47.

Summary

This PR introduces a professional JSON output mode for bdstorage, enabling seamless integration with automation pipelines, DevOps workflows, and external reporting systems.

Key Changes

Structured Output

• Added support for --output-format json.
• Produces a structured machine-readable summary of the deduplication run.

JSON Schema Support

The JSON output now includes:

• files_scanned — Total number of files analyzed.
• duplicate_groups — Number of duplicate file groups detected.
• bytes_saved — Actual disk space reclaimed during execution.
• vault_objects_added — Number of new objects added to the content-addressable vault.
• links_created — Total count of reflinks or hard links created.
• errors — Structured list of processing failures, including:
  • File path
  • Error reason
  • Relevant operation context

CI/CD Optimization

• Automatically disables interactive terminal UI elements in JSON mode, including:
  • Progress bars
  • Spinners
  • ANSI control sequences
• Ensures stdout remains clean and pipe-friendly for tools such as jq, CI runners, and log processors.

Improved Error Capture

• Refactored path-specific error handling so failures are collected as structured data rather than only printed to the terminal.
• Improves observability and downstream automation reliability.

Result

These enhancements make bdstorage significantly easier to integrate into larger automation ecosystems, backup workflows, monitoring pipelines, and enterprise storage tooling.