refactor(agent): rename md_to_html to readme_processor, fix section extraction, widget parsing, and list_base_models

- Rename md_to_html.py → readme_processor.py (file no longer just HTML conversion)
- _extract_section: include YAML frontmatter, use heading-level-aware forward
  walk (sub-headings under # are included), increase walk limit past 30 lines
- _is_heading: exclude </hN> closing tags from boundary detection
- _heading_level: new helper for heading-level-aware section matching
- css: yield 0 for heading like closing tags, was unexpectedly caught by _is_heading
- extract_gallery_images: fix YAML block scalar (text: >-) prompt extraction;
  use endswith instead of == to detect the block marker
- _strip_widget_section: add to clean_readme_for_llm (widget text is handled
  by post-processor, not needed in LLM prompt)
- _strip_standalone_images: keep markdown image URLs intact for LLM preview
  extraction (was stripping to alt text only)
- list_base_models: switch from scanner-cache aggregation to
  CivitaiBaseModelService.get_base_models() - always returns full list
- Ollama: add num_ctx=32768 to payload options so thinking models have room
  to both reason and produce output
- Add tests/agent_cli/test_readme_processor.py: 59 tests covering extraction,
  cleaning, section matching, heading detection
- Update existing tests for behavioral changes

py/services/llm_service.py

@@ -364,6 +364,9 @@ class LLMService:
                 "think": False,
                 "options": {
                     "temperature": temperature,
+                    # Allow up to 32K context so the model has room to think
+                    # AND produce output without hitting the 4K default limit.
+                    "num_ctx": 32768,
                 },
             }
             if response_format is not None:
@@ -381,6 +384,16 @@ class LLMService:
             if max_tokens is not None:
                 payload["max_tokens"] = max_tokens
 
+        if is_ollama:
+            logger.info(
+                "Ollama request: model=%s num_ctx=%s num_predict=%s format=%s think=%s",
+                payload.get("model"),
+                payload.get("options", {}).get("num_ctx"),
+                payload.get("options", {}).get("num_predict"),
+                payload.get("format", "none"),
+                payload.get("think"),
+            )
+
         headers = self._build_headers(cfg["api_key"])
 
         attempt = 0
@@ -507,8 +520,23 @@ class LLMService:
         )
 
         try:
-            return json.loads(result["content"])
+            parsed = json.loads(result["content"])
+            logger.info(
+                "LLM response base_model=%s tags=%s confidence=%s",
+                parsed.get("base_model", "?")[:50],
+                parsed.get("tags", []),
+                parsed.get("confidence", "?"),
+            )
+            logger.info(
+                "LLM raw content: %s",
+                (result.get("content") or "")[:1200],
+            )
+            return parsed
         except (json.JSONDecodeError, TypeError) as exc:
+            logger.info(
+                "LLM raw response (first 800 chars): %s",
+                (result.get("content") or "")[:800],
+            )
             logger.warning(
                 "LLM JSON parse failed on first attempt: %s. Retrying.", exc
             )