Fix network flows flaky test

[mariomac]

To speedup network flows testing, we were reusing the internal-pinger pod from a previous test, however, depending on the order of the events, the test might fail:

1. Internal-pinger is deployed.
2. The app o11y pipeline gets the internal-pinger metadata.
3. Internal pinger sends a ping.
4. Net o11y pipeline detects the packet before it gets the Pod metadata from the informer (it works in another thread than app o11y), so reports it without metadata.
5. The app o11y pipeline detects the call and sends it properly decorated.
6. The app o11y HTTP test succeeds and internal-pinger is removed.
7. The Net O11y integration test is run, assuming that the net pipeline should have detected the flow in the previous HTTP test.
8. The Net O11y integration test fails, because the flow was detected but not properly decorated with the internal-pinger metadata.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 80.21%. Comparing base (4f7e880) to head (6229a87).
Report is 2 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #1282      +/-   ##
==========================================
- Coverage   80.22%   80.21%   -0.01%     
==========================================
  Files         140      140              
  Lines       14058    14058              
==========================================
- Hits        11278    11277       -1     
+ Misses       2242     2241       -1     
- Partials      538      540       +2     

Flag	Coverage Δ
integration-test	59.26% <ø> (-0.02%)	⬇️
k8s-integration-test	57.71% <ø> (+0.02%)	⬆️
oats-test	35.25% <ø> (ø)
unittests	52.36% <ø> (-0.02%)	⬇️

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.

[grcevski]

LGTM!