Support StabilityAi Image Edit Api

[inpink]

Hello, this PR implements StabilityAi Image Edit - Remove Background API's client.

---

[The result of the execution]

Before	After

[Background]

StabilityAI provides various features for editing images via API requests.
Currently, Spring AI supports image generation using text, but lacks functionality for image editing.
This PR adds the StabilitAiEditApi class, which implements the Remove Background API from StabilityAI at the raw level.
The latest version of StabilityAI is V2 Beta, and for the API, please refer to StabilityAi Image Edit - Remove Background API.

[PR contents]

Component	Details
Request DTO	StabilityAiEditApi.RemoveBackgroundRequest record
Request Headers	StabilityAiEditApi.ImageEditHeaders record
Response DTO	Generic (StabilityAiEditApi.StructuredResponse record, byte[])
REST Call	removeBackground method in StabilityAiEditApi Class
Test	StabilityAiEditApiIT Class

• request:

Field	Type	Description
image	byte[]	A binary string representing the image from which users want to remove the background.
output_format	OutputFormat Enum	The format of the output image. PNG or WEBP.

Users can submit an image as byte[] in formats such as png or webp.
Raw binary image data can be submitted using multipart form data.
An HTTP request is sent using RestClient.
Using ImageEditHeaders, users can add headers. Headers added at the constructor level are included in the RestClient builder, while those added in the HTTP request sending method removeBackground are applied only to that specific request.

• response:

Response Type	Description
StructuredResponse	If a base64-encoded image is requested, the response is mapped to this class, containing the image, finish reason, and seed.
byte[]	If a raw binary image is requested, the response is mapped to a byte array containing the image data.

StructuredResponse Class

Field	Type	Description
b64Image	String	The generated image, encoded in base64 format.
finishReason	String	The reason the generation finished. Possible values are: - SUCCESS - CONTENT_FILTERED
seed	String	The seed used as random noise for this generation, which can help reproduce the result.

The response type can be set using the Accept header in the request.
Depending on the requested response type, it returns either a base64-encoded image or raw binary image data.
If a base64-encoded image is requested, the JSON object is mapped to the StructuredResponse class. If a raw binary image is requested, it is mapped to byte[]. Java Generics are used to dynamically determine the response type, which the user specifies in the responseType parameter when calling the removeBackground method.

• tests:
  Both tests ensure the functionality of the API in handling different response types (byte[] and StructuredResponse) and confirm that the data is processed and returned as expected.

1. removeBackgroundToRawImage Test:

Test Case	Description
Raw Binary Image Test	This test verifies that the API correctly processes raw binary image data. A sample image (test.png) is submitted in byte[] format.
Method Tested	removeBackground method
Expected Outcome	The test ensures that the method returns raw binary image data as a byte[].
Validation	The resulting image is saved and validated to confirm that it is not null.

2. removeBackgroundToBase64Image Test:

Test Case	Description
Base64-Encoded Image Test	This test checks if the API properly handles the request to return a base64-encoded image. A sample image (test.png) is submitted with the output format WEBP.
Method Tested	removeBackground method
Expected Outcome	The response is expected as a JSON object mapped to the StructuredResponse class, containing a base64-encoded image.
Assertions	The test asserts that the image, seed, and finish reason are present, and that the finish reason is "SUCCESS".
Validation	The base64-encoded image is decoded and saved as a WEBP file for further validation.

[Related Issue]

• Support Stable Diffusion #708 (help wanted labelled)

---

I’m pleased to submit this PR to Spring AI.
Please let me know if any revisions are needed, and I will address them promptly.
I would be happy to assist with future maintenance as well, if necessary.

[markpollack]

wow,thanks, getting to the PRs now, a nice addition! will merge soon.