fix(toc): add trailing bottom spacing when the TOC scrolls

The desktop TOC is its own scroll container (overflow-y:auto + max-height),
but unlike article content — which gets trailing space via
.page-content > article::after — it had nothing below the last link. A long
TOC scrolled to the bottom left the final link flush against the edge.

Add .toc::after as a block pseudo-element inside the desktop media query,
using the same technique (and the same Firefox-clips-padding-bottom reason)
as the article. Shared component, so public (about/charte/licence) and admin
(#admin-toc) both get it. Verified with Playwright: 0px -> ~60px gap;
mobile unaffected (TOC is not a scroll container there).

app/public/assets/css/components/toc.css

@@ -136,6 +136,15 @@
 		padding-right: var(--space-3xs);
 	}
 
+	/* Trailing bottom spacing when the TOC scrolls (long lists). Pseudo-element
+	   (real content in the scroll area) rather than the container's
+	   padding-bottom, which Firefox clips in this nested flex layout. */
+	.toc::after {
+		content: "";
+		display: block;
+		height: var(--space-xl);
+	}
+
 	.toc summary {
 		pointer-events: none;
 	}

app/src/MarkdownHelper.php

@@ -8,8 +8,11 @@ class MarkdownHelper
     /**
      * Extract h1–h3 headings from raw markdown content as TOC items.
      *
-     * Each heading gets an anchor id matching CommonMark's default slugification
-     * (lowercase, spaces → hyphens, punctuation stripped).
+     * The heading is parsed with CommonMark so inline markup (`*em*`,
+     * `**strong**`, `` `code` ``, `[links]()`) is resolved instead of leaking
+     * raw markdown into the TOC. The anchor id is slugified from the parsed
+     * inline text, which is what CommonMark uses to build the rendered heading
+     * ids — so TOC links and heading anchors always match.
      *
      * @return array<int, array{label: string, href: string, level: int}>
      */
@@ -17,19 +20,57 @@ class MarkdownHelper
     {
         $items = [];
         $lines = explode("\n", $content);
-
-
-        // Use CommonMark's own SlugNormalizer so TOC links match the rendered heading IDs exactly.
         $normalizer = new \League\CommonMark\Normalizer\SlugNormalizer();
 
         foreach ($lines as $line) {
-            if (preg_match('/^(#{1,3})\s+(.+)$/', $line, $m)) {
-                $level = strlen($m[1]);
-                $label = trim($m[2]);
-                $id = $normalizer->normalize($label);
-                $items[] = ['label' => $label, 'href' => '#' . $id, 'level' => $level];
+            if (!preg_match('/^(#{1,3})\s+(.+)$/', $line, $m)) {
+                continue;
             }
+
+            $headingText = self::parseInlineText(trim($m[2]));
+
+            $items[] = [
+                'label' => $headingText,
+                'href'  => '#' . $normalizer->normalize($headingText),
+                'level' => strlen($m[1]),
+            ];
         }
         return $items;
     }
+
+    /**
+     * Parse markdown and return its plain-text content.
+     *
+     * Walks the inline AST and concatenates the text nodes, so links keep their
+     * label (`[lien](url)` → "lien") and markup is dropped.
+     */
+    private static function parseInlineText(string $markdown): string
+    {
+        static $parser = null;
+        if ($parser === null) {
+            $environment = new \League\CommonMark\Environment\Environment([
+                'html_input' => 'strip',
+            ]);
+            $environment->addExtension(new \League\CommonMark\Extension\CommonMark\CommonMarkCoreExtension());
+            $parser = new \League\CommonMark\Parser\MarkdownParser($environment);
+        }
+
+        $document = $parser->parse($markdown);
+        $walker = $document->walker();
+        $text = '';
+
+        while ($event = $walker->next()) {
+            if (!$event->isEntering()) {
+                continue;
+            }
+            $node = $event->getNode();
+            if ($node instanceof \League\CommonMark\Node\Inline\Text
+                || $node instanceof \League\CommonMark\Extension\CommonMark\Node\Inline\Code
+            ) {
+                $text .= $node->getLiteral();
+            }
+        }
+
+        return trim($text);
+    }
 }