fix(sync): anchor device highlight ranges to selection text

KOReader highlights converted to wrong or unresolvable web CFIs
whenever a selection crossed text-node boundaries. Three compounding
defects: ParseCREXPointer discarded the text()[N] node index and
applied .offset as a global block offset; the end anchor converted
with an empty context (blind structural landing, zero verification);
and the block-containment check accepted wrong-node landings because
a selection's text always occurs somewhere in its own block. pos0 and
pos1 were converted independently, yielding mismatched start/end base
paths and backwards ranges - highlights stored but never painted, or
painted at the wrong location, in the web reader.

Add ConvertHighlightRange: both endpoints are anchored by locating the
selection text in the flattened document text (whitespace-normalized,
crossing block boundaries), with the device locators demoted to
disambiguation hints. A single document match yields a same-spine,
ordered start/end pair by construction. Falls back to the legacy
per-endpoint path when the text cannot be located, so behavior is
never worse than before.

Also fix the shared flatten stream: zero-width format characters
(ZWSP/ZWNJ/BOM) and whitespace-only inline spans (<span> </span>)
now act as word separators instead of fusing adjacent words, which
prevented text matching in real books.

Verified against 28 real device highlights from production: converter
output now matches the text-search oracle on all of them (previously
8/28 correct, 18 with mismatched base paths, 2 backwards).

internal/handlers/koreader.go

@@ -93,7 +93,22 @@ func (h *KOReaderHandler) convertHighlightPositions(ec annotationEpub, pos0, pos
 	if pos0 == "" || !ec.convertible() {
 		return "", ""
 	}
-	startLoc := wsync.ConvertToCanonical(wsync.LocatorSourceKOReader, pos0, percentage, contextText, ec.mediaItem.FormatGroup, ec.epubPath, "")
+	// Primary: anchor both endpoints by the selection text itself. One
+	// document match yields a same-spine ordered pair — the invariant the
+	// web reader's range-CFI builder needs. Independent per-endpoint
+	// conversion mis-anchors any selection past a block's first text node
+	// (crengine text-node indices don't survive to the raw XHTML).
+	startLoc := wsync.ConvertHighlightRangeToCanonical(ec.epubPath, ec.mediaItem.FormatGroup, pos0, pos1, contextText, percentage)
+	if startLoc.Precision == "exact" || startLoc.Precision == "structural" {
+		startCFI := webUsableCFI(startLoc)
+		endCFI := ""
+		if startCFI != "" && startLoc.EndCFI != "" {
+			endCFI = webUsableCFI(wsync.CanonicalLocator{CFI: startLoc.EndCFI, Precision: startLoc.Precision})
+		}
+		return startCFI, endCFI
+	}
+	// Fallback: legacy independent per-endpoint conversion.
+	startLoc = wsync.ConvertToCanonical(wsync.LocatorSourceKOReader, pos0, percentage, contextText, ec.mediaItem.FormatGroup, ec.epubPath, "")
 	startCFI := webUsableCFI(startLoc)
 	endCFI := ""
 	// A text-search start matched the selection text itself: its extent