variables component agent

[isthaison]

What problem does this PR solve?

Add variable component agent
Proceduce:

UI:

agent test:

abc.json

Briefly describe what this PR aims to solve. Include background context that will help reviewers understand the purpose of the PR.

Type of change

• Bug Fix (non-breaking change which fixes an issue)
• New Feature (non-breaking change which adds functionality)
• Documentation Update
• Refactoring
• Performance Improvement
• Other (please describe):

[isthaison]

Hi @KevinHuSh , as you know the request changed. I tried to optimize it. so that the canvas code does not change too much.

[Sci-fiBrain]

This Feature is great. Why hasn't it been merged yet? Are there any problems being solved?

[isthaison]

I don't know either, I'm not on their team.

[coco2006]

if the "Code" component is there, you could use the "Code" snippet to handle the similar scenario.

[isthaison]

With the above scenario is simple, but with more complex scenarios, the code component can not handle it.
With this component, the purpose is to use llm to extract information into json, then extract json into the variable of begin. so that the components can use these variables accordingly, here are some examples:
Case 1: Extract normal information to pre-define variables from begin, described initially.
Case 2: Extract dynamic information according to the input value passed in, suppose: there are 10 formdata to process for different actions (this data is stored in retrial , {"type": "action 1", "param1":"" } ,{"type": "action 2", "param2":"" } ), combined with the conversation between the user and AI, to extract. then based on the type to branch for the agent according to each type of action

[KevinHuSh]

Appreciations!
We're gona refactor this part a lot recently.