Add dividend discount model (DDM) skill to financial-analysis vertical

[HumphreySun98]

What

Adds a new ddm-model skill to the financial-analysis vertical — a Dividend Discount Model (income-approach) valuation skill, and the income-based complement to the existing dcf-model and comps-analysis skills.

Why

The financial-analysis vertical covers DCF, trading comps, LBO, and 3-statement models, but there is no income / dividend-based valuation method anywhere in the repo. The DDM is the standard approach precisely where an FCF-DCF is unreliable or ill-defined:

• Banks and insurers, where "free cash flow" isn't meaningful
• REITs, utilities, and mature high-payout companies whose primary return is the distribution
• As an income-based cross-check on a DCF or trading-comps valuation

It does not overlap dcf-model: a DDM discounts dividends at the cost of equity (not WACC) and yields equity value per share directly — no unlevered FCF and no enterprise-to-equity bridge. The skill hammers this cost-of-equity-not-WACC distinction, which is the most common DDM error.

What's inside (mirrors the dcf-model gold-standard file pattern)

• SKILL.md — MCP-first data-source priority; the "Formulas Over Hardcodes" and verify-step-by-step constraints; the DDM workflow; a model-selection guide (single- / two- / three-stage / H-model); correct patterns & common mistakes; Excel structure; and a decision-support / human-sign-off guardrail.
• references/methodology.md — single-, two-, and three-stage math and the H-model closed form; the ROE-driven build for financials (g = ROE · retention, book-value roll-forward); total-payout (buyback) handling; justified P/E and P/B cross-checks; DCF/comps reconciliation; and a fully worked two-stage example.
• scripts/validate_ddm.py — an openpyxl validator that mirrors validate_dcf.py's structure and JSON output. It checks formula errors, enforces the critical cost of equity > terminal growth constraint, and flags an out-of-range cost of equity, an aggressive terminal growth, a terminal value that dominates total value, and a payout ratio above 100%. Exit 0 = PASS.
• TROUBLESHOOTING.md and requirements.txt (openpyxl only).

Validation

• python3 scripts/check.py → OK — 80 file(s) checked, 0 issues.
• scripts/validate_ddm.py was tested against clean and deliberately broken workbooks (PASS/exit 0 and FAIL/exit 1), plus edge cases: a 0% (no-growth) perpetuity, per-share vs aggregate value bases, and vertical assumption layouts.
• The worked two-stage example's arithmetic was verified independently (intrinsic value ≈ $42.52; terminal value 77.1% of total).
• The pre-commit hook auto-bumped financial-analysis plugin.json 0.1.1 → 0.1.2. The new skill is not bundled into any agent, so there is no sync drift and no other manifest change.

Frontmatter is name + description only, matching the existing vertical skills. This PR is self-contained and touches only the new skill directory plus the automatic version bump.