fix: error in agent mode while using models not supporting tools

[sinomoe]

Description

In the "ReAct Agent" mode, after executing a tool, a "ToolPromptMessage" is saved. Since the model in "ReAct" mode definitely does not support function calling, it is unable to calculate the token count of the "ToolPromptMessage." The fix is to determine whether to calculate the token count of the "ToolPromptMessage" based on whether the model's feature supports "tool calling."

Fixes #5055

Type of Change

Please delete options that are not relevant.

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• This change requires a documentation update, included: Dify Document
• Improvement, including but not limited to code refactoring, performance optimization, and UI/UX improvement
• Dependency upgrade

How Has This Been Tested?

Please describe the tests that you ran to verify your changes. Provide instructions so we can reproduce. Please also list any relevant details for your test configuration

• manual test

Suggested Checklist:

• I have performed a self-review of my own code
• I have commented my code, particularly in hard-to-understand areas
• My changes generate no new warnings
• I ran dev/reformat(backend) and cd web && npx lint-staged(frontend) to appease the lint gods
• optional I have made corresponding changes to the documentation
• optional I have added tests that prove my fix is effective or that my feature works
• optional New and existing unit tests pass locally with my changes

[sinomoe]

It seems like that this is a common issue in models which do not support tools calling. there is the same issue with volcengine maas models. Perhaps we should avoid calculating the tool prompt's token count in agent mode while using models that do not support tools calling?

[sinomoe]

found another way to fix the issue, refer to #5245