chore: improve gemini models

[hjlarry]

Summary

Tip

Close issue syntax: Fixes #<issue number> or Resolves #<issue number>, see documentation for more details.

1. add some missing features of some models

2. use the property of MultiModelPromptMessageContent

3. fix may raise error on windows platform

2024-12-16 23:14:12,294.294 ERROR [Thread-11 (_generate_worker)] [app_generator.py:239] - Error when generating
Traceback (most recent call last):
  File "C:\Users\hejl\PycharmProjects\dify\api\core\model_runtime\model_providers\__base\large_language_model.py", line 110, in invoke
    result = self._invoke(
             ^^^^^^^^^^^^^
  File "C:\Users\hejl\PycharmProjects\dify\api\core\model_runtime\model_providers\google\llm\llm.py", line 66, in _invoke
    return self._generate(model, credentials, prompt_messages, model_parameters, tools, stop, stream, user)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\hejl\PycharmProjects\dify\api\core\model_runtime\model_providers\google\llm\llm.py", line 195, in _generate
    content = self._format_message_to_glm_content(msg)
              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\hejl\PycharmProjects\dify\api\core\model_runtime\model_providers\google\llm\llm.py", line 383, in _format_message_to_glm_content
    glm_content["parts"].append(self._upload_file_content_to_google(c))
                                ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\hejl\PycharmProjects\dify\api\core\model_runtime\model_providers\google\llm\llm.py", line 365, in _upload_file_content_to_google
    os.unlink(temp_file.name)
PermissionError: [WinError 32] 另一个程序正在使用此文件，进程无法访问。: 'C:\\Users\\hejl\\AppData\\Local\\Temp\\tmpear9twnj'

Screenshots

Before	After
...	...

Checklist

Important

Please review the checklist below before submitting your pull request.

• This change requires a documentation update, included: Dify Document
• I understand that this PR may be closed in case there was no previous discussion or issues. (This doesn't apply to typos!)
• I've added a test for each change that was introduced, and I tried as much as possible to make a single atomic change.
• I've updated the documentation accordingly.
• I ran dev/reformat(backend) and cd web && npx lint-staged(frontend) to appease the lint gods