Expand DOPEY into a full product presentation

[Chris-ObtuseAI]

What changed

• replaced the compressed public README with a 1,326-line product presentation;
• explained the complete evidence lifecycle and symbol-decision workflow;
• documented the capability atlas across data, technicals, macro, fundamentals, options, targeting, squeeze, bubble, risk, truth, memory, autonomy, desktop, and governed execution;
• added detailed squeeze and bubble engine mechanics, default research thresholds, and explicit non-claims;
• documented per-symbol null testing, overlap collapse, chronological era gates, and candidate generations;
• documented recursive improvement, protected mutation surfaces, rollback, transactional snapshots, and dynamic scheduling;
• published the complete 39-stage supervisor inventory;
• added eight architecture and workflow diagrams plus sixteen evidence/capability tables;
• preserved the approved Evidence Before Promotion hero and public clean-slate boundary;
• removed mojibake by keeping the presentation ASCII-safe.

Why

The prior README was a polished summary, but it did not fully explain how DOPEY works, what its subsystems can do, or why its evidence-first architecture is differentiated. The repository landing surface needed to function as a complete technical product presentation.

Impact

Visitors can now understand the full system from the repository landing page without reverse-engineering packages or historical reports. Claims remain tied to code, versioned policies, validation evidence, or explicit limitations.

Validation

• GitHub Markdown API render accepted the README
• 59 local links resolved
• 8 Mermaid blocks present
• strict UTF-8 and ASCII-safe scan passed
• python -m operator_tools.verify_public_source_boundary
• pytest -q tests/test_text_encoding_guard.py - 5 passed
• pip check
• git diff --check