Move PcapLiveDevice statistics machinery entirely inside the .cpp file.

[Dimi1010]

Split of #1838, since that one has been sitting for a while.

This part moves only the statistics worker to be entirely internal to the PcapLiveDevice translation unit.

[codecov]

Codecov Report

❌ Patch coverage is 66.66667% with 12 lines in your changes missing coverage. Please review.
✅ Project coverage is 83.48%. Comparing base (96484fb) to head (0b08957).
⚠️ Report is 1 commits behind head on dev.

Files with missing lines	Patch %	Lines
Pcap++/src/PcapLiveDevice.cpp	66.66%	9 Missing and 3 partials ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##              dev    #1966      +/-   ##
==========================================
- Coverage   83.51%   83.48%   -0.03%     
==========================================
  Files         311      311              
  Lines       54897    54881      -16     
  Branches    12213    12182      -31     
==========================================
- Hits        45845    45820      -25     
- Misses       7788     8207     +419     
+ Partials     1264      854     -410     

Flag	Coverage Δ
alpine320	75.95% <52.17%> (-0.01%)	⬇️
fedora42	76.08% <54.16%> (-0.03%)	⬇️
macos-13	81.64% <69.23%> (-0.01%)	⬇️
macos-14	81.64% <69.23%> (-0.01%)	⬇️
macos-15	81.65% <69.23%> (-0.01%)	⬇️
mingw32	70.34% <41.66%> (+0.01%)	⬆️
mingw64	70.33% <41.66%> (+0.13%)	⬆️
npcap	?
rhel94	75.80% <52.17%> (-0.01%)	⬇️
ubuntu2004	60.24% <57.14%> (-0.04%)	⬇️
ubuntu2004-zstd	60.36% <57.14%> (+<0.01%)	⬆️
ubuntu2204	75.74% <52.17%> (-0.03%)	⬇️
ubuntu2204-icpx	60.73% <42.30%> (-0.03%)	⬇️
ubuntu2404	75.99% <52.17%> (-0.02%)	⬇️
ubuntu2404-arm64	75.60% <52.17%> (-0.03%)	⬇️
unittest	83.48% <66.66%> (-0.03%)	⬇️
windows-2022	85.48% <60.00%> (+0.12%)	⬆️
windows-2025	85.51% <64.28%> (+0.09%)	⬆️
winpcap	85.51% <64.28%> (-0.10%)	⬇️
xdp	53.57% <0.00%> (+0.01%)	⬆️

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

☔ View full report in Codecov by Sentry.
📢 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.

[seladb]

This PR reverts many of the changes done here: #1785, I'm curious why we're moving back and forth? 🤔

[Dimi1010]

This PR reverts many of the changes done here: #1785, I'm curious why we're moving back and forth? 🤔

The first change was an attempt to separate the variables only needed for the thread procedures away from the rest of the device.

As I was reworking the capture thread variables, I figured that having said variables grouped as contexts and passed to a pure function procedure would be better for internal representation. That way, the level of indirection done by the polymorphic Worker class is removed while also keeping the encapsulation of the thread variables used only inside the procedure.

Since the workers are one-offs and won't be reused, it is a simpler solution that would only keep std::thread and std::atomic<bool> inside the device class.
Additionally, it would allow reuse of the stop flag, as the same flag can be passed to multiple procedures. In the worker scenario, each worker created its own stop flag.