SEO polish: OG image, favicons, 404, article schema, image optimization

- Add branded 1200x630 og-default.jpg; article pages use their hero as
  og:image, others use the default
- Wire the existing (previously unlinked) favicon set into every page
- Add branded 404.html with root-absolute paths + noindex
- Emit Article + localized BreadcrumbList JSON-LD on all article pages
- Convert 74 heavy served photos to WebP (28MB -> 8.5MB, -70%), rewrite
  references, add loading=lazy/decoding=async to below-fold images and
  fetchpriority=high to heroes (protect LCP)
- Stop tracking images/original/ (458MB of unused source masters; kept on
  disk, gitignored) and ignore .playwright-mcp/

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

tools/seo_inject.py

@@ -10,7 +10,7 @@ Usage:
     python seo_inject.py --all      # write every page
     python seo_inject.py --all --sitemap   # also (re)build sitemap + robots
 """
-import os, re, sys, html
+import os, re, sys, html, json
 try:
     sys.stdout.reconfigure(encoding="utf-8", errors="replace")
 except Exception:
@@ -27,11 +27,33 @@ LANGS = ["en", "mk", "si"]
 HREF = {"en": "en", "mk": "mk", "si": "sl"}
 OG_LOCALE = {"en": "en_GB", "mk": "mk_MK", "si": "sl_SI"}
 
+# Section (first path segment) display names, for Article breadcrumbs.
+# Any 2-segment path "section/slug" whose section is here is treated as an article.
+SECTION_NAMES = {
+    "blog": {"en": "Blog", "mk": "Блог", "si": "Blog"},
+    "news": {"en": "News", "mk": "Новости", "si": "Novice"},
+    "projects": {"en": "Events & projects", "mk": "Настани и проекти", "si": "Dogodki in projekti"},
+    "get-to-know-slovenia": {"en": "Get to know Slovenia", "mk": "Запознај ја Словенија", "si": "Spoznaj Slovenijo"},
+    "get-to-know-macedonia": {"en": "Get to know Macedonia", "mk": "Запознај ја Македонија", "si": "Spoznaj Makedonijo"},
+    "student-welcome-guide": {"en": "Student Welcome Guide", "mk": "Водич за добредојде", "si": "Vodič za bruce"},
+}
+HOME_CRUMB = {"en": "Home", "mk": "Дома", "si": "Domov"}
+
 START, END = "<!-- SEO:START -->", "<!-- SEO:END -->"
 BLOCK_RE = re.compile(re.escape(START) + r".*?" + re.escape(END) + r"\s*", re.DOTALL)
 TITLE_RE = re.compile(r"(<title>.*?</title>)", re.DOTALL)
 
-DEFAULT_OG_IMAGE = DOMAIN + "/images/1-logo.png"
+DEFAULT_OG_IMAGE = DOMAIN + "/images/og-default.jpg"
+
+# Favicons are ACTUAL resources the browser loads for the current host, so they
+# stay root-relative (resolve on the temp host, local server, and final domain).
+FAVICON_LINES = [
+    '    <link rel="icon" type="image/png" href="/favicon-96x96.png" sizes="96x96">',
+    '    <link rel="icon" type="image/svg+xml" href="/favicon.svg">',
+    '    <link rel="shortcut icon" href="/favicon.ico">',
+    '    <link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">',
+    '    <link rel="manifest" href="/site.webmanifest">',
+]
 
 # ---- page trees (for hreflang parity) ----
 def build_trees():
@@ -115,8 +137,8 @@ def extract_description(content, lang=None, path=None):
     return None
 
 HERO_PATTERNS = [
-    r'<figure class="article-main-image">.*?<img src="(.*?)"',
-    r'<figure class="news-article-hero">\s*<img src="(.*?)"',
+    r'<figure class="article-main-image">.*?<img[^>]*?src="(.*?)"',
+    r'<figure class="news-article-hero">\s*<img[^>]*?src="(.*?)"',
 ]
 
 def extract_og_image(content):
@@ -136,9 +158,46 @@ def title_text(content):
         return SITE_NAME["en"]
     return clean_text(m.group(1))
 
+H1_RE = re.compile(r"<h1[^>]*>(.*?)</h1>", re.DOTALL)
+def extract_h1(content):
+    m = H1_RE.search(content)
+    return clean_text(m.group(1)) if m else None
+
 def canonical_for(lang, path):
     return f"{DOMAIN}/{lang}/" + (f"{path}/" if path else "")
 
+def article_schema(lang, path, content, canon, desc, ogimg):
+    """Article + BreadcrumbList JSON-LD for 2-segment 'section/slug' pages."""
+    parts = path.split("/")
+    if len(parts) != 2 or parts[0] not in SECTION_NAMES:
+        return []
+    section, _ = parts
+    headline = extract_h1(content) or title_text(content).split(" - ")[0]
+    org = {"@type": "Organization", "name": SITE_NAME[lang], "url": f"{DOMAIN}/{lang}/"}
+    article = {
+        "@context": "https://schema.org", "@type": "Article",
+        "headline": headline,
+        "description": desc or "",
+        "image": ogimg,
+        "inLanguage": HREF[lang],
+        "author": org,
+        "publisher": {"@type": "Organization", "name": SITE_NAME[lang],
+                      "logo": {"@type": "ImageObject", "url": f"{DOMAIN}/images/1-logo.png"}},
+        "mainEntityOfPage": canon,
+    }
+    if not article["description"]:
+        del article["description"]
+    crumbs = {
+        "@context": "https://schema.org", "@type": "BreadcrumbList",
+        "itemListElement": [
+            {"@type": "ListItem", "position": 1, "name": HOME_CRUMB[lang], "item": f"{DOMAIN}/{lang}/"},
+            {"@type": "ListItem", "position": 2, "name": SECTION_NAMES[section][lang], "item": f"{DOMAIN}/{lang}/{section}/"},
+            {"@type": "ListItem", "position": 3, "name": headline, "item": canon},
+        ],
+    }
+    dump = lambda o: '    <script type="application/ld+json">' + json.dumps(o, ensure_ascii=False, separators=(",", ":")) + "</script>"
+    return [dump(article), dump(crumbs)]
+
 def build_block(lang, path, content):
     canon = canonical_for(lang, path)
     title = title_text(content)
@@ -190,6 +249,8 @@ def build_block(lang, path, content):
         '</script>'
     )
     lines.append(org)
+    lines.extend(article_schema(lang, path, content, canon, desc, ogimg))
+    lines.extend(FAVICON_LINES)
     lines.append("    " + END)
     return "\n".join(lines)