CI: Expose configuration to ignore parent E2E expect test runner invocations

[michaeldiamant]

Summary

Extends scripts/buildtools/check_tests.py to customize tests to ignore via --ignored-tests. The motivation is to acquiesce check_tests.py warnings about E2E expect tests run multiple times.

In another branch, I saw warnings in https://app.circleci.com/pipelines/github/algorand/go-algorand/8787/workflows/a566ab26-42ad-48de-940d-08d504ce1c15/jobs/158801?invite=true#step-103-7. Reproduced below for convenience:

=========== RAN MULTIPLE TIMES ===================
The above 4 tests ran multiple times:
github.com/algorand/go-algorand/test/e2e-go/cli/algod/expect TestAlgodWithExpect -- ran 10 times. (Can probably be fixed by adding "partitiontest.PartitionTest()")
github.com/algorand/go-algorand/test/e2e-go/cli/algoh/expect TestAlgohWithExpect -- ran 10 times. (Can probably be fixed by adding "partitiontest.PartitionTest()")
github.com/algorand/go-algorand/test/e2e-go/cli/goal/expect TestGoalWithExpect -- ran 10 times. (Can probably be fixed by adding "partitiontest.PartitionTest()")
github.com/algorand/go-algorand/test/e2e-go/cli/tealdbg/expect TestTealdbgWithExpect -- ran 10 times. (Can probably be fixed by adding "partitiontest.PartitionTest()")
The above 4 tests ran multiple times:
==================================================

Upon inspection, I realized these tests are parent E2E test runners for expect tests. They're invoked multiple times by-design and the advice to add partitioning does not apply.

As a band aid, the PR exposes a way to ignore these tests. See the PR's builds for proof that the warnings disappear.

• Treat the PR optionally. I didn't see a way to prevent the tests from reporting via go test flags. I can envision managing the ignore list as worse than the status quo.
• I realize the verification step happens for all test types. The configuration relies on test name uniqueness.
• For the PR's scope, I'm gauging if the band aid is deemed a net positive. If it isn't, I'd prefer to close and move on.

Test Plan

Run CI.

[codecov]

Codecov Report

Merging #4422 (baf8f59) into master (f156f69) will decrease coverage by 0.05%.
The diff coverage is n/a.

@@            Coverage Diff             @@
##           master    #4422      +/-   ##
==========================================
- Coverage   55.18%   55.13%   -0.06%     
==========================================
  Files         397      397              
  Lines       50073    50073              
==========================================
- Hits        27634    27606      -28     
- Misses      20150    20170      +20     
- Partials     2289     2297       +8     

Impacted Files	Coverage Δ
ledger/roundlru.go	90.56% <0.00%> (-5.67%)	⬇️
util/db/dbutil.go	44.24% <0.00%> (-4.25%)	⬇️
ledger/blockqueue.go	85.63% <0.00%> (-2.88%)	⬇️
crypto/merkletrie/trie.go	66.42% <0.00%> (-2.19%)	⬇️
crypto/merkletrie/node.go	91.62% <0.00%> (-1.87%)	⬇️
cmd/tealdbg/debugger.go	72.69% <0.00%> (-0.81%)	⬇️
ledger/acctupdates.go	69.92% <0.00%> (-0.61%)	⬇️
network/wsPeer.go	67.67% <0.00%> (-0.55%)	⬇️
ledger/acctonline.go	78.47% <0.00%> (-0.53%)	⬇️
network/wsNetwork.go	64.89% <0.00%> (+0.19%)	⬆️
... and 1 more

📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more

[jannotti]

Good find.