refactor(sdk-client): replace ConfigurationImpl class with createConfiguration factory function

[joker23]

Requirements

• I have added test coverage for new or changed functionality
• I have followed the repository's pull request submission guidelines
• I have validated my changes against all supported platform versions

Related issues

SDK-1705

Describe the solution you've provided

This PR replaces the ConfigurationImpl class with a createConfiguration() factory function following the CONTRIBUTING.md guidelines for preferring interfaces over classes. The factory function returns a Configuration interface object instead of instantiating a class, which results in better bundle size and minification potential.

Key changes:

• Replaced ConfigurationImpl class with createConfiguration() factory function in Configuration.ts
• Validation logic is inlined directly in the factory function for reduced bundle size
• Uses object spread to construct the return value, excluding internal URI properties
• Updated LDClientImpl to use createConfiguration() instead of new ConfigurationImpl()
• Updated exports in configuration/index.ts to export createConfiguration (removed ConfigurationImpl)
• Updated all test files to use createConfiguration() and verify properties through the public Configuration interface

Important for reviewers:

1. The test assertions for baseUri, eventsUri, streamUri were changed to check serviceEndpoints.polling, serviceEndpoints.streaming, serviceEndpoints.events respectively - these internal URI properties are not part of the public Configuration interface
2. Similarly, payloadFilterKey tests now check serviceEndpoints.payloadFilterKey since that's where the value is stored in the public interface
3. ConfigurationImpl was removed entirely - this is a breaking change if any external consumers were using it directly (internal usage only expected)

Bundle size impact:

• js-client-sdk-common: -762 bytes raw (-0.77%), -89 bytes brotli (-0.46%)
• js-client-sdk: -196 bytes raw (-0.24%), +4 bytes brotli (+0.02%)

Human review checklist:

• Verify validation logic behavior is preserved (type coercion for booleans, minimum value enforcement, error logging)
• Confirm no external consumers depend on ConfigurationImpl export
• Check that object spread doesn't leak internal properties (baseUri, eventsUri, streamUri, payloadFilterKey are destructured out)

Describe alternatives you've considered

Could have kept ConfigurationImpl for backwards compatibility, but since it's internal-only usage, removing it entirely provides cleaner code and better bundle size.

Additional context

Link to Devin run: https://app.devin.ai/sessions/608a317f9f334f38ab018c4b83ed3b4c

Requested by: Steven Zhang (@joker23)

---

Note

Medium Risk
Touches core config construction/validation and exported API surface; behavior should be equivalent but any subtle differences in option coercion or unknown-option handling could affect SDK initialization.

Overview
Refactors SDK configuration creation by removing the ConfigurationImpl class and introducing a createConfiguration() factory that builds a plain Configuration object, inlining option validation/coercion and constructing ServiceEndpoints while omitting internal URI fields from the returned shape.

Updates LDClientImpl and unit tests to use createConfiguration(), and adjusts assertions to validate endpoint defaults and payloadFilterKey via config.serviceEndpoints rather than internal baseUri/streamUri/eventsUri properties; updates configuration/index.ts exports accordingly.

^{Written by Cursor Bugbot for commit f194310. This will update automatically on new commits. Configure here.}

[devin-ai-integration]

🤖 Devin AI Engineer

I'll be helping with this pull request! Here's what you should know:

✅ I will automatically:

• Address comments on this PR. Add '(aside)' to your comment to have me ignore it.
• Look at CI failures and help fix them

Note: I can only respond to comments from users who have write access to this repository.

⚙️ Control Options:

• Disable automatic comment and CI monitoring

[github-actions]

@launchdarkly/js-sdk-common size report
This is the brotli compressed size of the ESM build.
Compressed size: 25394 bytes
Compressed size limit: 26000
Uncompressed size: 124693 bytes

[github-actions]

@launchdarkly/js-client-sdk-common size report
This is the brotli compressed size of the ESM build.
Compressed size: 18896 bytes
Compressed size limit: 20000
Uncompressed size: 97324 bytes

[github-actions]

@launchdarkly/js-client-sdk size report
This is the brotli compressed size of the ESM build.
Compressed size: 23232 bytes
Compressed size limit: 25000
Uncompressed size: 80673 bytes

[github-actions]

@launchdarkly/browser size report
This is the brotli compressed size of the ESM build.
Compressed size: 171357 bytes
Compressed size limit: 200000
Uncompressed size: 797784 bytes