feat(translation): quality pipeline overhaul + new features (audit 2026-08-29)

Translation quality & format preservation:
- Word: merge adjacent same-format runs into one unit (sentence-level
  coherence like inline-tag handling); translate comments/balloons;
  dedupe textbox collection (was translated twice); RTL no longer
  overrides center/justify alignment; CJK/Arabic font hints (eastAsia/cs)
- PPTX: chart translations now actually reach the output file
  (ChartPart.blob is read-only — rewrite chart XML in the saved ZIP);
  CJK typeface hints (a:ea)
- Excel: sheet renames no longer break references — rewrite cell
  formulas (3D/quoted), defined names, data validations, cond. formats
- PDF: bold/italic honored (hebo/heit/hebi); table cells never merge;
  unchanged blocks left untouched (typography preserved, fixes duplicate
  hyperlinks); attempted/changed stats + route gate now cover PDF;
  CJK font paths; scanned PDFs via Mistral OCR (detection + admin settings)

Features:
- formality param (formal/informal) + automatic regional-variant prompts
- output_mode=bilingual docx (source above translation)
- per-user translation memory on Redis (falls back to LRU), context-hashed
- QA report + 0-100 confidence score in job status; L0 on by default
- OpenAI-compatible providers: whole chunk in ONE numbered-JSON request
  (~15x fewer calls) with per-item fallback; base prompt always present
  (custom prompt no longer replaces translation instructions)

Infra & marketing alignment:
- plan-based engine gating + vision gating (closes paid-engine leak);
  /providers/available filtered per plan; 107 languages exposed
- zh-CN/zh-TW validation fixed; libmagic disabled on Windows (native crash)
- admin: Mistral OCR settings + engine status dashboard; httpx<0.28 pin
  (TestClient breakage); Prometheus test fixture fixed
- marketing docs aligned with code (PDF+OCR, retention, engines, pricing)
- security: .env.ionos/.env.production/provider_settings.json removed

Tests: 1173 passed / 0 failed (6 network tests deselected: free Google
endpoint temporarily blocked from this machine)

services/glossary_service.py

@@ -198,9 +198,11 @@ def build_full_prompt(
     source_lang: str = "fr",
     target_lang: str = "en",
     glossary_target_lang: str = "multi",
+    formality: Optional[str] = None,
 ) -> str:
     """
-    Build the complete prompt combining custom prompt and glossary.
+    Build the complete prompt combining custom prompt, glossary, formality
+    and regional variant directives.
 
     Args:
         custom_prompt: Optional custom system prompt from user
@@ -208,6 +210,8 @@ def build_full_prompt(
         source_lang: ISO code of the source language
         target_lang: ISO code of the target language
         glossary_target_lang: ISO code of the glossary's target language configuration
+        formality: Optional tone override — "formal" or "informal". Only
+            meaningful for LLM engines (ignored by classic engines).
 
     Returns:
         Combined prompt string
@@ -224,4 +228,30 @@ def build_full_prompt(
         if glossary_prompt:
             parts.append(glossary_prompt)
 
+    if formality in ("formal", "informal"):
+        if formality == "formal":
+            parts.append(
+                "TONE: Use a formal, professional register throughout "
+                "(formal address (vous/Sie) where the language distinguishes; "
+                "no slang, no contractions where avoidable)."
+            )
+        else:
+            parts.append(
+                "TONE: Use an informal, natural register throughout "
+                "(tu-style address where the language distinguishes; "
+                "contractions welcome)."
+            )
+
+    # Regional variant: when the target code carries a region (pt-BR,
+    # fr-CA, zh-CN...), make the expected variety explicit — LLMs default
+    # to the dominant variant otherwise (pt-PT, fr-FR...).
+    if target_lang and "-" in target_lang and target_lang != "auto":
+        from core.languages import language_name
+
+        name = language_name(target_lang)
+        if name and name != target_lang:
+            parts.append(
+                f"REGIONAL VARIANT: write specifically in {name}."
+            )
+
     return "\n\n".join(parts) if parts else ""