feat(market-research): add Xquik social listening and align scoring

[kriptoburak]

Summary

• Add optional, read-only X social-listening collection through Xquik MCP or REST.
• Preserve query provenance, timestamps, post URLs, engagement fields, pagination state, and sampling limits.
• Treat public posts as untrusted convenience-sample data and separate evidence from interpretation.
• Align /analyze-feedback with the sentiment-analysis Skill score range of -1 to +1.
• Rebuild the branch on current main and add the required CHANGELOG.md credit.

Independent Fix

The command report template used an X/10 average while its underlying Skill required -1 to +1. The first signed commit fixes that mismatch and adds a regression test for the shared score contract.

Xquik Integration

The existing pm-market-research plugin now supports Xquik as an optional source for current public X research. The core Skill stays concise and loads a dedicated collection guide only when needed. That guide covers MCP and REST setup, bounded queries, opaque cursor pagination, field normalization, failure handling, research-quality safeguards, and exact source attribution. No X write endpoint is used.

Validation

• python3 validate_plugins.py
• python3 -m unittest discover -s tests -v (16 tests)
• claude plugin validate pm-market-research --strict
• claude plugin validate . --strict
• Skill, JSON, shell-example, JavaScript-example, link, description-alignment, and secret-pattern checks
• Public Xquik OpenAPI verification for search, pagination parameters, authentication, and tweet lookup
• git diff --check

All executable checks pass locally. The GitHub Tests run awaits maintainer approval (action_required) and created 0 jobs. The marketplace still contains 9 plugins, 68 skills, and 42 commands. Versions remain at 2.1.0; contributor guidance leaves version bumps to release time.

Xquik is an independent third-party service. Not affiliated with X Corp. "Twitter" and "X" are trademarks of X Corp.