Mobile expand-panel icon doesn't bump to enlarged size

[jacuzzicoding]

Summary

The inline expand panel (ItemExpandPanel) icon is invisible on mobile. Desktop renders the bumped 192px (and 128px at ≤1180px) correctly per PR #99, but at ≤720px the icon does not appear at any size.

Expected

The hand-drawn icon shows at a meaningful size on mobile — either the desktop 192/128 stepping continues into the mobile range, or a dedicated mobile size lands somewhere between row size (48) and the 128 step.

Actual

The icon is hidden entirely at ≤720px. The expand panel renders without an icon.

Repro

1. Open the app at a viewport ≤720px (mobile, narrow window, or DevTools device emulation).
2. Open any town's category tab (e.g. fish).
3. Expand a row whose item has a hand-drawn icon committed (e.g. fish/sea-bass or fish/koi on an ACGCN town).
4. Observe: the expand panel shows MonthGrid + stats, but no icon.

Suspected cause

src/index.css line 1528–1530 hides the icon on mobile via:

@media (max-width: 720px) {
  .ac-expand { padding-left: 18px; }
  .ac-expand-icon { display: none; }
}

This rule is pre-existing — it was introduced when <ItemIcon> first landed in v0.9.1 (commit bdaea9a) and was preserved in PR #99 alongside the new desktop responsive steps. The original motivation was that the 64px icon competed with the MonthGrid + stats stack on a narrow single-column layout. With hand-drawn art now treated as a specimen-card centerpiece, hiding it on mobile is at odds with the design intent.

So this is not a PR #99 regression — it is a pre-existing CSS choice that the new aesthetic makes wrong.

Suggested fix

Remove the display: none rule and pick a mobile-appropriate size. Two options:

• Continue the 192 → 128 cascade with a 96 (or 80) step at ≤720px, keeping the icon absolute-positioned and adjusting padding-left accordingly. Lower-risk; small CSS delta.
• Restructure the panel on mobile so the icon flows inline above the MonthGrid (top-of-card hero treatment) rather than absolute-positioned in the corner. Cleaner at narrow widths but a larger change.

The first option is the lower-risk path; the second is the better long-term answer if mobile becomes more central.