docs: add RAG failure analysis guide (WFGY 16-problem map)

[onestardao]

Closes #719

Pull Request Checklist

• I have read and followed the CONTRIBUTING.md guidelines.
• My contribution adds a new instruction, prompt, agent, skill, or workflow file in the correct directory.
• The file follows the required naming convention.
• The content is clearly structured and follows the example format.
• I have tested my instructions, prompt, agent, skill, or workflow with GitHub Copilot.
• I have run npm start and verified that README.md is up to date.

---

Description

This PR adds a new instruction file:

instructions/wfgy-core-reasoning-os.instructions.md

It provides a plain-text reasoning core (WFGY Core 2.0) designed to be used as a custom GitHub Copilot instruction.

The goal is to offer a minimal, infrastructure-free reasoning layer that improves multi-step reasoning stability for tasks such as:

• debugging and refactoring
• architectural reasoning
• structured analysis
• complex pipeline design

The file includes:

• A drop-in system prompt block
• A short self-test section for quick validation
• Clear structure and usage guidance

This contribution was discussed in:
#719 (comment)

Happy to adjust formatting, naming, or structure if needed.

---

Type of Contribution

• New instruction file.
• New prompt file.
• New agent file.
• New plugin.
• New skill file.
• New agentic workflow.
• Update to existing instruction, prompt, agent, plugin, skill, or workflow.
• Other (please specify):

---

Additional Notes

This instruction is intentionally lightweight and does not introduce any external dependencies or workflow changes.
It is designed to be purely text-based and compatible with standard GitHub Copilot custom instruction usage.

---

By submitting this pull request, I confirm that my contribution abides by the Code of Conduct and will be licensed under the MIT License.

[aaronpowell]

Why isn't everything after this just the instructions file? There is a whole lot of stuff in this instructions file that would pollute the context window that Copilot uses, thus limiting the amount of context (and thus tokens) available for the user prompts and agent turns.