feat(schema): improve HTML preprocessing for schema generation

Add new preprocess_html_for_schema utility function to better handle HTML cleaning
for schema generation. This replaces the previous optimize_html function in the
GoogleSearchCrawler and includes smarter attribute handling and pattern detection.

Other changes:
- Update default provider to gpt-4o
- Add DEFAULT_PROVIDER_API_KEY constant
- Make LLMConfig creation more flexible with create_llm_config helper
- Add new dependencies: zstandard and msgpack

This change improves schema generation reliability while reducing noise in the
processed HTML.

pyproject.toml

@@ -42,7 +42,9 @@ dependencies = [
     "pyperclip>=1.8.2",
     "faust-cchardet>=2.1.19",
     "aiohttp>=3.11.11",
-    "humanize>=4.10.0"
+    "humanize>=4.10.0",
+    "zstandard>=0.23.0",
+    "msgpack>=1.1.0"
 ]
 classifiers = [
     "Development Status :: 4 - Beta",