fix(tradein-avito): strip Emotion CSS from listings.address

Avito DOM leaks inline <style>.css-XXX{...}</style> inside the location <p>,
which gets pulled by .text(strip=True). Add _clean_address() helper that strips
CSS rules + post-address noise (Площадь N года, от N мин.), apply at extraction
site. Also applied to address_full in avito_detail.py via import from avito.py.
Backfill migration cleans the ~hundreds of historical rows.

Source: estimate a0a0b820-e8a8-4eee-aa73-0ab3b98ac233 analog #4.

tradein-mvp/backend/app/services/scrapers/avito.py

@@ -205,7 +205,7 @@ class AvitoScraper(BaseScraper):
             if loc_el is not None:
                 street_el = loc_el.css_first("p")
                 if street_el is not None:
-                    address = street_el.text(strip=True) or None
+                    address = _clean_address(street_el.text(strip=True))
 
             # Фото
             photo_urls: list[str] = []
@@ -267,6 +267,29 @@ _RE_STUDIO = re.compile(r"студи[яиюей]", re.IGNORECASE)
 _RE_AREA = re.compile(r"(\d+[.,]?\d*)\s*м[²2]", re.IGNORECASE)
 _RE_FLOOR = re.compile(r"(\d+)\s*/\s*(\d+)\s*эт\.?", re.IGNORECASE)
 
+_CSS_NOISE_RE = re.compile(r"\.?css-[a-z0-9_-]+\s*\{[^}]*\}", flags=re.I)
+_NOT_ADDRESS_TAIL_RE = re.compile(
+    r"\s*(Площадь \d|от \d+\s?мин\.|css-[a-z0-9_-]+)",
+    flags=re.I,
+)
+
+
+def _clean_address(raw: str | None) -> str | None:
+    """Strip Emotion CSS and post-address noise that Avito DOM leaks via .text().
+
+    Examples:
+        "ул. Токарей, 56к1Площадь 1905 года.css-39hgr0{fill:..."
+            -> "ул. Токарей, 56к1"
+        "ул. Малышева, 1.css-xxx{...}"
+            -> "ул. Малышева, 1"
+    """
+    if not raw:
+        return None
+    cleaned = _CSS_NOISE_RE.sub("", raw)
+    cleaned = _NOT_ADDRESS_TAIL_RE.split(cleaned, maxsplit=1)[0]
+    cleaned = cleaned.strip(" ,.\n\t")
+    return cleaned or None
+
 
 def _extract_rooms_from_title(title: str) -> int | None:
     if _RE_STUDIO.search(title):

tradein-mvp/backend/app/services/scrapers/avito_detail.py

@@ -32,6 +32,7 @@ from selectolax.parser import HTMLParser, Node
 from sqlalchemy import text
 from sqlalchemy.orm import Session
 
+from app.services.scrapers.avito import _clean_address
 from app.services.scrapers.avito_exceptions import AvitoBlockedError, AvitoRateLimitedError
 
 logger = logging.getLogger(__name__)
@@ -297,7 +298,7 @@ def parse_detail_html(html: str, source_url: str) -> DetailEnrichment:
         lon = _try_float(map_el.attributes.get("data-map-lon"))
         avito_location_id = _try_int(map_el.attributes.get("data-location-id"))
 
-    address_full = _text(tree, "[itemprop='address']")
+    address_full = _clean_address(_text(tree, "[itemprop='address']"))
 
     # ── Description ─────────────────────────────────────────────────────────
     description = _text(tree, "[data-marker='item-view/item-description']")

tradein-mvp/backend/data/sql/062_clean_avito_addresses.sql

@@ -0,0 +1,15 @@
+-- 062_clean_avito_addresses.sql
+-- Strip CSS noise from listings.address for Avito source (one-shot backfill).
+-- Idempotent: re-run is safe (regexp_replace on already-clean string is a no-op).
+BEGIN;
+
+UPDATE listings
+   SET address = trim(both ' ,.' FROM
+                       regexp_replace(
+                         regexp_replace(address, '\.?css-[a-z0-9_-]+\s*\{[^}]*\}', '', 'gi'),
+                         '\s*(Площадь \d|от \d+\s?мин\.|css-[a-z0-9_-]+).*$', '', 'i'
+                       ))
+ WHERE source = 'avito'
+   AND (address ~* 'css-[a-z0-9_-]+' OR address ~* 'Площадь \d|от \d+\s?мин\.');
+
+COMMIT;

tradein-mvp/backend/tests/test_avito_address_clean.py

@@ -0,0 +1,24 @@
+from app.services.scrapers.avito import _clean_address
+
+
+def test_strips_css_noise():
+    raw = (
+        "ул. Токарей, 56к1Площадь 1905 года.css-39hgr0{fill:currentColor;"
+        "height:1em;box-sizing:border-box;}.css-39hgr0:focus{outline:none;}"
+        "от 31 мин.Геологическая.css-39hgr0{...}"
+    )
+    assert _clean_address(raw) == "ул. Токарей, 56к1"
+
+
+def test_clean_address_passthrough():
+    assert _clean_address("ул. Малышева, 1") == "ул. Малышева, 1"
+
+
+def test_clean_handles_none():
+    assert _clean_address(None) is None
+    assert _clean_address("") is None
+    assert _clean_address("   ") is None
+
+
+def test_clean_strips_trailing_punctuation():
+    assert _clean_address("ул. Ленина, 5,") == "ул. Ленина, 5"