fix(sdk-trace-base): fix spanLimits attribute length/count to consider env values

[svetlanabrennan]

Which problem is this PR solving?

OTEL_ATTRIBUTE_VALUE_LENGTH_LIMIT and OTEL_ATTRIBUTE_COUNT_LIMIT env values were being ignored when re-configuring span limits.

Fixes #3049

Short description of the changes

Updated the reconfigureLimits function to check additional items before updating spanLimits.attributeValueLengthLimit and spanLimits.attributeCountLimit property.

BEFORE
Before, reconfigureLimits wasn't considering if the user set the general attribute value length limit (OTEL_ATTRIBUTE_VALUE_LENGTH_LIMIT) via env - it was only checking if user set that value programmatically. Same issue with the attribute count limit OTEL_ATTRIBUTE_COUNT_LIMIT.

NOW
Now it checks if the user did not set the spanLimits.attributeValueLengthLimit programmatically AND if OTEL_SPAN_ATTRIBUTE_VALUE_LENGTH_LIMIT env value is still set to the default value (ex. Infinity) BUT if the user set the generalLimits.attributeValueLengthLimit programmatically OR user set the OTEL_ATTRIBUTE_VALUE_LENGTH_LIMIT via env, then reconfigure the spanLimits.attributeValueLengthLimit value to either use the generalLimits.attributeValueLengthLimit set programmatically or use the OTEL_ATTRIBUTE_VALUE_LENGTH_LIMIT env value.

Note: I had to wrap the old DEFAULT_CONFIG object into a function because those envs value (parsed in DEFAULT_CONFIG) were being parsed before the tests ran so it caused the new tests to fail. I added a comment about that above that function just in case someone had questions about it but I can remove the comment if needed.

Type of change

Please delete options that are not relevant.

• Bug fix (non-breaking change which fixes an issue)

How Has This Been Tested?

Please describe the tests that you ran to verify your changes. Provide instructions so we can reproduce. Please also list any relevant details for your test configuration

• Unit tests

Checklist:

• Followed the style guidelines of this project
• Unit tests have been added

[codecov]

Codecov Report

Merging #3068 (86fc4f6) into main (776656b) will decrease coverage by 0.44%.
The diff coverage is 88.00%.

❗ Current head 86fc4f6 differs from pull request most recent head b959b72. Consider uploading reports for the commit b959b72 to get more accurate results

@@            Coverage Diff             @@
##             main    #3068      +/-   ##
==========================================
- Coverage   93.11%   92.67%   -0.45%     
==========================================
  Files         189      174      -15     
  Lines        6276     5555     -721     
  Branches     1319     1181     -138     
==========================================
- Hits         5844     5148     -696     
+ Misses        432      407      -25     

Impacted Files	Coverage Δ
...ckages/opentelemetry-core/src/utils/environment.ts	90.90% <40.00%> (-5.10%)	⬇️
...elemetry-sdk-trace-base/src/BasicTracerProvider.ts	95.49% <100.00%> (ø)
...ackages/opentelemetry-sdk-trace-base/src/config.ts	87.17% <100.00%> (+0.33%)	⬆️
...ckages/opentelemetry-sdk-trace-base/src/utility.ts	100.00% <100.00%> (ø)
packages/opentelemetry-resources/karma.worker.js	0.00% <0.00%> (-100.00%)	⬇️
...kages/opentelemetry-sdk-trace-base/karma.worker.js	0.00% <0.00%> (-100.00%)	⬇️
...lemetry-resources/src/detectors/BrowserDetector.ts	53.33% <0.00%> (-46.67%)	⬇️
...lemetry-resources/src/detectors/ProcessDetector.ts	95.45% <0.00%> (-4.55%)	⬇️
...telemetry-sdk-trace-web/src/StackContextManager.ts
...ckages/opentelemetry-sdk-trace-web/karma.worker.js
... and 13 more

[pichlermarc]

Looks good, thanks for working on this! 🙂
reconfigureLimits() is a bit hard to read, but maybe some comments could help 🤔

[legendecas]

I don't think this is a "chore" patch. IMO, it's more like a "feat" or "fix", from various perspectives.

[legendecas]

Overall LGTM. Thank you for your contribution!