[ISSUE] workload_size should expect a str, not an object which references value

[longchass]

Description
A clear and concise description of what the bug is.
Same as Issue #494 and #495

For some reason, this is not merged and is still reflected on the SDK docs

[https://databricks-sdk-py.readthedocs.io/en/latest/dbdataclasses/serving.html#databricks.sdk.service.serving.ServingEndpointDetailed](link to Docs)

Reproduction
A minimal code sample demonstrating the bug.
ServedModelInput(
model_name=model_name,
model_version=latest_model_version,
workload_size="Small",
scale_to_zero_enabled=True,
)
Expected behavior
A clear and concise description of what you expected to happen.
The above code should pass based on the documentations, but instead fails because it expects and ENUM value

Is it a regression?
Did this work in a previous version of the SDK? If so, which versions did you try?
No, current latest version that is being used on Databricks Workspace notebook

Debug Logs
The SDK logs helpful debugging information when debug logging is enabled. Set the log level to debug by adding logging.basicConfig(level=logging.DEBUG) to your program, and include the logs here.

Other Information

• OS: [e.g. macOS]
• Version: [e.g. 0.1.0]

Additional context
Add any other context about the problem here.