feat: two-phase pre-screening with HTTP check + DeepSeek batch classification

Phase 1 (no AI credits): httpx checks every selected domain concurrently
(30 parallel) with real browser UA — detects live/dead/parked/redirect.
Parked: keyword scan in body/title + known parking host redirect check.
Results saved to DB immediately; dead/parked never reach DeepSeek.

Phase 2 (single DeepSeek call): all live-site titles + snippets bundled
into ONE Replicate/DeepSeek-R1 request → returns niche + type for every
domain in batch (up to 80 per call, parallelised if more).

- app/prescreener.py (new): _check_one(), prescreen_domains(),
  classify_with_deepseek(), parking signal lists, same-domain redirect logic
- app/db.py: prescreen_status/niche/site_type/prescreen_at columns +
  migrations; save_prescreen_results() upsert helper
- app/main.py: POST /api/prescreen/batch endpoint
- app/static/index.html:
  - 🔍 Pre-screen button (disabled while running, shows spinner)
  - Niche + Type columns in Browse and Leads tables (.pni/.pty pills)
  - Prescreen status colour dot (●) when niche not yet set
  - prescreening state flag; result toast shows per-status counts

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

app/main.py

@@ -18,7 +18,7 @@ from app.db import (
     DATA_DIR, PARQUET_PATH, SQLITE_PATH,
     init_db, get_stats, get_domains, get_enriched,
     queue_domains, get_queue_status, build_duckdb_index, index_status,
-    queue_ai, get_ai_queue_status, save_ai_assessment,
+    queue_ai, get_ai_queue_status, save_ai_assessment, save_prescreen_results,
 )
 from app.enricher import start_worker, pause_worker, resume_worker, is_running, ensure_workers_alive
 from app.scorer import run_scoring
@@ -171,6 +171,56 @@ async def enriched(
 
 # ── AI assessment endpoints ───────────────────────────────────────────────────
 
+@app.post("/api/prescreen/batch")
+async def prescreen_batch(body: dict):
+    """
+    Phase 1 — HTTP check every domain (no AI).  Marks live/dead/parked/redirect.
+    Phase 2 — Single DeepSeek call for all live domains → niche + type.
+    Max 200 domains per call.
+    """
+    domains = body.get("domains", [])
+    if not domains:
+        return JSONResponse({"error": "no domains provided"}, status_code=400)
+    if len(domains) > 200:
+        return JSONResponse({"error": "max 200 domains per batch"}, status_code=400)
+
+    from app.prescreener import prescreen_domains, classify_with_deepseek, DEEPSEEK_BATCH_SIZE
+
+    # Phase 1: HTTP checks (concurrent, no AI)
+    results = await prescreen_domains(domains)
+    await save_prescreen_results(results)
+
+    counts: dict = {}
+    for r in results:
+        s = r.get("prescreen_status", "dead")
+        counts[s] = counts.get(s, 0) + 1
+
+    # Phase 2: DeepSeek classification for live sites only
+    live = [r for r in results if r.get("prescreen_status") == "live"]
+    classified = 0
+    if live:
+        batches = [live[i:i + DEEPSEEK_BATCH_SIZE] for i in range(0, len(live), DEEPSEEK_BATCH_SIZE)]
+        batch_cls = await asyncio.gather(
+            *[classify_with_deepseek(b) for b in batches], return_exceptions=True
+        )
+        all_cls: list = []
+        for bc in batch_cls:
+            if isinstance(bc, list):
+                all_cls.extend(bc)
+        if all_cls:
+            await save_prescreen_results(all_cls)
+            classified = len(all_cls)
+
+    return {
+        "total":      len(domains),
+        "live":       counts.get("live",     0),
+        "parked":     counts.get("parked",   0),
+        "redirect":   counts.get("redirect", 0),
+        "dead":       counts.get("dead",     0),
+        "classified": classified,
+    }
+
+
 @app.post("/api/ai/assess/batch")
 async def ai_assess_batch(body: dict):
     domains_list = body.get("domains", [])