fix(scrapy): make logging configuration idempotent

[vdusek]

initialize_logging re-wrapped scrapy.utils.log.configure_logging on every call, stacking wrappers when it ran more than once. The monkey-patch is now installed at most once and reads the current handler and level from a module-level state dict, so repeated initialization stays correct without nesting wrappers.

Part of splitting the larger Scrapy integration fix (fix/scrapy-integration) into reviewable pieces.

[codecov]

Codecov Report

❌ Patch coverage is 14.28571% with 12 lines in your changes missing coverage. Please review.
✅ Project coverage is 86.88%. Comparing base (db9444f) to head (80112be).
⚠️ Report is 1 commits behind head on master.

Files with missing lines	Patch %	Lines
src/apify/scrapy/_logging_config.py	14.28%	12 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #954      +/-   ##
==========================================
- Coverage   87.08%   86.88%   -0.20%     
==========================================
  Files          48       48              
  Lines        2973     2982       +9     
==========================================
+ Hits         2589     2591       +2     
- Misses        384      391       +7     

Flag	Coverage Δ
e2e	36.51% <0.00%> (-0.12%)	⬇️
integration	58.21% <0.00%> (-0.18%)	⬇️
unit	75.31% <14.28%> (-0.17%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[Pijukatel]

Maybe I am missing some context, but can't we skip the module-level dict and do it like this?

@lru_cache
def _configure_all_loggers(handler, level):
    if handler is None:
        return
    for logger_name in [None, *_ALL_LOGGERS]:
        _configure_logger(logger_name, level, handler)

[vdusek]

Unfortunately, we can't do that - the _configure_all_loggers needs to run more than once.

The expected flow of the Scrapy Actors is something like this:

1. We configure logging (our handler + levels).
2. We do some setup, during which things may already log.
3. Scrapy's own setup runs and overrides our logging configuration.
4. We re-apply our configuration on top to override it back.

So the current code is the correct way.