Try to_dict method when encoding JSON

[alexmojaki]

With this code:

import vertexai
from vertexai.generative_models import GenerativeModel

import logfire

logfire.configure()
vertexai.init()
model = GenerativeModel('gemini-1.5-flash-002')
chat = model.start_chat()
response = chat.send_message('What is 2+2?')
logfire.info('response', response=response)

this PR changes the attributes from:

{
  "response": "candidates {\n  content {\n    role: \"model\"\n    parts {\n      text: \"2 + 2 = 4\\n\"\n    }\n  }\n  finish_reason: STOP\n  avg_logprobs: -5.3347812354331836e-06\n}\nusage_metadata {\n  prompt_token_count: 7\n  candidates_token_count: 8\n  total_token_count: 15\n}\nmodel_version: \"gemini-1.5-flash-002\"\n"
}

to

{
  "response": {
    "candidates": [
      {
        "content": {
          "role": "model",
          "parts": [
            {
              "text": "2 + 2 = 4\n"
            }
          ]
        },
        "finish_reason": "STOP",
        "avg_logprobs": -0.000005319886440702248
      }
    ],
    "usage_metadata": {
      "prompt_token_count": 7,
      "candidates_token_count": 8,
      "total_token_count": 15
    },
    "model_version": "gemini-1.5-flash-002"
  }
}

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 100.00%. Comparing base (6ab0661) to head (42df0bc).
Report is 1 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff            @@
##              main      #799   +/-   ##
=========================================
  Coverage   100.00%   100.00%           
=========================================
  Files          138       138           
  Lines        11036     11040    +4     
  Branches      1565      1565           
=========================================
+ Hits         11036     11040    +4     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.