[Feature]: SAM CLI durable emulator integration tests are ported to this repository

### What would you like?

Context: Changes to the `aws-durable-execution-sdk-python` and `aws-durable-execution-sdk-python-testing` can often break assertions in the SAM CLI integration tests:

1. https://github.com/aws/aws-sam-cli/pull/8906
2. https://github.com/aws/aws-sam-cli/actions/runs/28089573325/job/83163914199

At the time of writing these tests in SAM CLI, the intention was to test the emulator behaviour. Now that we automatically publish an emulator image for SAM CLI from this repository, I think we can loosen some of the tests on SAM CLI side. SAM CLI doesn't need to test the specifics of our SDK behaviour, only that durable functions can be run and that the execution history can be fetched in SAM CLI. The specific behaviour of the published emulator image in conjunction with the Python SDK (such as the presence of certain fields in the execution history) can be tested in this repository rather than in SAM CLI.

We want to have stronger ownership over our SAM CLI integration as well as reduce churn for SAM CLI owners.

### Possible Implementation

1. Check emulator image test coverage in current repository
2. Check emulator image test coverage in SAM CLI
3. Reason over tests we can keep in SAM and tests we can port (define how decoupled we want to be between the durable emulator image and SAM CLI, decide minimum functionality we want to test on SAM side)
4. Port over relevant tests to current repository that we are missing
5. Remove relevant tests from SAM CLI
### Is this a breaking change?

No

### Does this require an RFC?

No

### Additional Context

This is really similar to https://github.com/aws/aws-durable-execution-sdk-python/issues/430. We could consider porting these tests and running them via the SAM CLI for this issue as well.

首先，将此存储库中的模拟器镜像测试覆盖率与 SAM CLI 集成测试进行比较，并参考链接的 SAM CLI pull request 和 workflow 运行。定义需要保留的最低 SAM CLI 行为，将此处缺失的模拟器特定覆盖率移植过来，并验证在不丢失 durable-function 执行和历史记录覆盖率的情况下可以减少 SAM CLI 测试。

由索引模型根据 Issue 内容生成。