[#1286] feat(frontend): add AMM vs SDEX depth comparison view in LiquidityFragmentation page

[ToryMic]

Summary

Adds the AMM vs SDEX venue split that #1286 asks for on the Liquidity Fragmentation page. The page previously showed total liquidity as one aggregated number, with no breakdown between Stellar Classic order book depth and Soroban AMM pool reserves.

This consumes the existing GET /api/v1/assets/:symbol/liquidity endpoint via the existing useLiquidity hook. That endpoint already returns per-venue bidDepth / askDepth / priceLevels for SDEX and StellarX AMM, and no frontend code was using it on this page.

What was added

1. Interactive stacked area chart (VenueDepthComparison) — cumulative depth with one stacked area per venue, a bid/ask side toggle, hover tooltips, and a legend that labels each venue as limit order book or pooled reserves.
2. Venue slippage calculator (VenueSlippageCalculator) — estimated price impact for 10k / 50k / 100k USDC orders on each venue, so the cheapest venue for a given notional is visible at a glance.
3. venueModel.ts / venueSlippage.ts — pure helpers for venue-model classification, cumulative depth curve construction, and slippage maths. Kept out of the components so they are directly testable.

Notes on the modelling

• SDEX depth is a discrete limit order book, so depth grows as price moves away from the mid.
• Soroban AMM liquidity is pooled reserves, so it is flat across impact levels.

Plotting both on one cumulative curve is what makes the two models comparable. Where a venue cannot absorb an order, the calculator reports insufficient depth rather than a misleadingly small percentage, and flags that cross-venue routing would be needed.

Testing

• 36 new tests: venueModel.test.ts, VenueSlippageCalculator.test.ts, VenueDepthComparison.test.tsx, LiquidityFragmentation.test.tsx
• All 36 pass
• eslint clean on all changed files
• tsc --noEmit reports no errors in any changed file

Pre-existing baseline failures (not touched by this PR): the frontend suite on main already fails with 76 failing tests across 15 files (useAlertSnoozes, useBridgeNotes, HealthScoreCard, AssetDetail, and the admin token-storage tests, among others). Before and after this branch that count is identical — 15 failed files / 76 failed tests — so this change adds no new failures. .github/workflows/ci.yml and docker.yml also currently fail to parse on main (${{ runner.temp }} in job-level env, and ${{ id.meta.* }}), so the CI jobs for this PR may not report. Both are upstream issues and I have left them alone here; happy to fix them in a separate PR if useful.

Reviewer notes

• No backend changes were needed — the venue-split data was already available and unused.
• The new components sit alongside the existing LiquidityDepthChart / PriceImpactCalculator, which operate on the combined book. Mine are per-venue, so the two are complementary rather than duplicative.

Closes #1286

[drips-wave]

@ToryMic Great news! 🎉 Based on an automated assessment of this PR, the linked Wave issue(s) no longer count against your application limits.

You can now already apply to more issues while waiting for a review of this PR. Keep up the great work! 🚀

Learn more about application limits