fix: Make percentage jumps use byte-weighted section bounds

Jumping to a percentage of the book picked a chapter by walking the
capped position list in equal steps, which no longer matched how
reading progress itself is reported. In a book with one huge chapter
and many small ones, that mismatch could send a 50% jump to a small
chapter near the end instead of the middle of the chapter that
actually holds half the content. It now picks the chapter the same
way progress is reported, by size, so jumping to a percentage and
reading toward it land on the same book position.

internal/webui/static/reader-engine.js

@@ -204,8 +204,18 @@ export class ReaderEngine extends HTMLElement {
       if (finite(target.index)) return target;
       if (finite(target.fraction)) {
         const fraction = Math.max(0, Math.min(1, target.fraction));
-        const position = this.positionList[Math.min(this.positionList.length - 1, Math.floor(fraction * this.positionList.length))];
-        return { index: this.book.sections.findIndex(s => s.id === position.href), locator: position };
+        // Use the same byte-weighted section bounds relocate() reports
+        // progress with, not the position list's raw index spacing: with
+        // a bounded fallback list, a chapter's share of positions can be
+        // far smaller than its share of the book's bytes, and indexing
+        // straight into positionList would then favor small chapters
+        // over the one the reader actually meant to reach.
+        const bounds = this.getSectionFractions();
+        let index = bounds.length - 2;
+        while (index > 0 && bounds[index] > fraction) index--;
+        const span = bounds[index + 1] - bounds[index];
+        const anchor = span > 0 ? (fraction - bounds[index]) / span : 0;
+        return { index, anchor };
       }
       if (target.href) return { index: this.book.sections.findIndex(s => s.id === target.href.split("#")[0]), locator: Locator.deserialize(target) };
     }