Support sending multipart/form-data in RestProxy

[jianghaolu]

• Add a @PartParam and parse it in SwaggerMethodParser
• Send multipart/form-data in RestProxy

[alzimmermsft]

@jianghaolu do you have any specific service scenarios this could be tested against to verify correctness?

[jianghaolu]

Most cognitive services accept multipart/form-data as input, e.g.: https://github.com/Azure/azure-rest-api-specs/blob/2e7ca21a5ad6f749e5bd453580f1c31c1d99b4e6/specification/cognitiveservices/data-plane/ComputerVision/stable/v1.0/ComputerVision.json#L563.

[alzimmermsft]

The implementation of multipart/form-data should follow the specification laid out in RFC 7578. Based on initial review of the specification the following is known:

Option Name	Required	Default (if not required)	Explanation
Name	Yes	N/A	Name of the form field, configures name in the part Content-Disposition header
Filename	No	None	Name of a file included in the form, configures filename in the part Content-Disposition header
ContentType	No	text/plain	The Content-Type of the part, if the part is a file this will default to application/octet-stream

Depending on the usage of multipart/form-data new wrapper types may need to be created as all values above can be determined at runtime.

[weidongxu-microsoft]

loadtesting uses multipart, and they plan a GA soon (expect Nov.)
#30706

At present they uses a helper method to use setHeader and setBody to provide a solution (the case when streaming is not necessary?).
https://github.com/Azure/azure-sdk-for-java/pull/30706/files#diff-5b7425837268825f2f8d2712d0ce73ade89309dc7ec407d2c9a45708525af0f9
And provide a hand-written method in Client to use this helper.

Let us know your opinion of this solution.

[weidongxu-microsoft]

We may need an update on it. OpenAI would use multipart/form-data in mid Sept. for transcription from video to text

[github-actions]

Hi @jianghaolu, we deeply appreciate your input into this project. Regrettably, this issue has remained inactive for over 2 years, leading us to the decision to close it. We've implemented this policy to maintain the relevance of our issue queue and facilitate easier navigation for new contributors. If you still believe this topic requires attention, please feel free to create a new issue, referencing this one. Thank you for your understanding and ongoing support.