test(dag/walker): fix flaky bloom FP assertion

[lidel]

Allow up to 2 false positives instead of exactly 0 in the default-rate regression test. At 1 in 4.75M with 100K probes the expected count is ~0.02, giving a ~2% flake rate for an exact-zero assertion.

I think the purpose of the test should be to warn us if suddenly something gets totally out of whack, and it should not make our live miserable by being flaky, so dialing this a notch down should be a good compromise.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 62.99%. Comparing base (a696a54) to head (0656e97).
⚠️ Report is 1 commits behind head on main.

@@            Coverage Diff             @@
##             main    #1139      +/-   ##
==========================================
- Coverage   63.00%   62.99%   -0.01%     
==========================================
  Files         266      266              
  Lines       26661    26661              
==========================================
- Hits        16797    16795       -2     
- Misses       8149     8150       +1     
- Partials     1715     1716       +1     

see 7 files with indirect coverage changes

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.