fix: map token usage metadata for Anthropic Claude model

[hemasekhar-p]

Link to Issue or Description of Change

1. Link to an existing issue (if applicable):

• Closes: Claude (Anthropic) model wrapper drops token usage — Event.usageMetadata() is always empty #1259

2. Or, if no issue exists, describe the change:

Problem:
The Anthropic Claude model wrapper in the ADK Java currently droping the token usage information returned by the Anthropic API. Because of this, LlmResponse.usageMetadata() returns Optional.empty() for Claude-backed agents.

Solution:
I have updated convertAnthropicResponseToLlmResponse in Claude.java to extract inputTokens and outputTokens from the Anthropic Message.usage() payload. The values are mapped to GenerateContentResponseUsageMetadata

Testing Plan

Unit Tests:

• I have added or updated unit tests for my change.
• All unit tests pass locally.

Added testClaudeUsageMapping_ShouldMapTokensCorrectly to ClaudeTest.java. The test verifies that inputTokens, outputTokens, and totalTokenCount are correctly mapped from the Anthropic SDK response to the ADK UsageMetadata builder.

Checklist

• I have read the CONTRIBUTING.md document.
• My pull request contains a single commit.
• I have performed a self-review of my own code.
• I have commented my code, particularly in hard-to-understand areas.
• I have added tests that prove my fix is effective or that my feature works.
• New and existing unit tests pass locally with my changes.
• I have manually tested my changes end-to-end.

[hemasekhar-p]

@tilgalas, Could you please review this PR.

[rogerzxu]

This is the same implementation I used in my source-forked copy and confirmed it works, so LGTM

[tilgalas]

thanks for the PR - left some minor comments, but overall LGTM