perf: offload BeautifulSoup parsing to a thread via asyncio.to_thread (#1817)

## Summary

- `BeautifulSoup()` constructor performs synchronous HTML parsing that
blocks the event loop, degrading concurrency.
- Wrapped `parse()` and `parse_text()` in `asyncio.to_thread()` to
offload parsing to a worker thread, consistent with how
`ParselParser.parse()` already works.

---

Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Author: vdusek

src/crawlee/crawlers/_beautifulsoup/_beautifulsoup_parser.py

@@ -1,5 +1,6 @@
 from __future__ import annotations
 
+import asyncio
 from typing import TYPE_CHECKING, Literal
 
 from bs4 import BeautifulSoup, Tag
@@ -23,11 +24,12 @@ def __init__(self, parser: BeautifulSoupParserType = 'lxml') -> None:
 
     @override
     async def parse(self, response: HttpResponse) -> BeautifulSoup:
-        return BeautifulSoup(await response.read(), features=self._parser)
+        body = await response.read()
+        return await asyncio.to_thread(BeautifulSoup, body, features=self._parser)
 
     @override
     async def parse_text(self, text: str) -> BeautifulSoup:
-        return BeautifulSoup(text, features=self._parser)
+        return await asyncio.to_thread(BeautifulSoup, text, features=self._parser)
 
     @override
     def is_matching_selector(self, parsed_content: Tag, selector: str) -> bool: