Enhancement Podcast Generator tool to support all TTS models

[bsormagec]

Summary

This pull request enhances the Podcast Generator tool to support all TTS models, removing the dependency on OpenAI API and allowing for a more flexible and comprehensive text-to-speech functionality. This change simplifies the configuration process and broadens the tool's usability.

Fixes #12887

Screenshots

Before	After

[crazywoola]

Please link an existing issue or open an issue first.

[bsormagec]

Please link an existing issue or open an issue first.

I added it. Thanks for letting me know.

[bsormagec]

If you need anything, feel free to reach out! I really appreciate the fantastic work everyone has done on Dify. @crazywoola @laipz8200

[laipz8200]

Those changes will break the workflow using this tool. Please wait for our plugin function release, then the tool will be under version control.

[bsormagec]

Those changes will break the workflow using this tool. Please wait for our plugin function release, then the tool will be under version control.

If you'd like, I can introduce this tool as a new tool, under a different name, or ensure backward compatibility. This flexibility enhances its usability.

[laipz8200]

It's a good idea to make it a new tool, but we don't accept new tools now. Please wait for the v1.0.0 version release. It will not be too long.

[bsormagec]

It's a good idea to make it a new tool, but we don't accept new tools now. Please wait for the v1.0.0 version release. It will not be too long.

Please tag or add something to this PR for v1, and we can move forward once it's ready.

[crazywoola]

FYI, you need to submit plugins here https://github.com/langgenius/dify-official-plugins
These changes only applies for version under v1.0, we will release v1.0 soon. You might need to move these changes to this repo https://github.com/langgenius/dify-official-plugins/tree/main/models