Load and save the lightGBM classifier not producing the same results

Nobody has claimed this yet.

### SynapseML version

0.10.2

### System information

- **Language version** (e.g. python 3.8, scala 2.12): python 3.8
- **Spark Version** (e.g. 3.2.3): 3.3.2
- **Spark Platform** (e.g. Synapse, Databricks): Databricks

### Describe the problem

Trained a lightGBM classifier. I tried to save this in the native format and load it back again with the below code -

`LightGBMClassificationModel.loadNativeModelFromString(model.getNativeModel())`

I am not getting the same results with the load and save model as compared to original.

Can someone please help?

### Code to reproduce issue

```python
LightGBMClassificationModel.loadNativeModelFromString(model.getNativeModel())
```

### Other info / logs

_No response_

### What component(s) does this bug affect?

- [ ] `area/cognitive`: Cognitive project
- [ ] `area/core`: Core project
- [ ] `area/deep-learning`: DeepLearning project
- [ ] `area/lightgbm`: Lightgbm project
- [ ] `area/opencv`: Opencv project
- [ ] `area/vw`: VW project
- [ ] `area/website`: Website
- [ ] `area/build`: Project build system
- [ ] `area/notebooks`: Samples under notebooks folder
- [ ] `area/docker`: Docker usage
- [ ] `area/models`: models related issue

### What language(s) does this bug affect?

- [ ] `language/scala`: Scala source code
- [ ] `language/python`: Pyspark APIs
- [ ] `language/r`: R APIs
- [ ] `language/csharp`: .NET APIs
- [ ] `language/new`: Proposals for new client languages

### What integration(s) does this bug affect?

- [ ] `integrations/synapse`: Azure Synapse integrations
- [ ] `integrations/azureml`: Azure ML integrations
- [ ] `integrations/databricks`: Databricks integrations

Start by reproducing the discrepancy with the provided Python call to LightGBMClassificationModel.loadNativeModelFromString(model.getNativeModel()) in the stated SynapseML, Spark, and Databricks versions. Compare predictions before and after native serialization, then trace the corresponding LightGBM model save/load entry points; done means the loaded model produces the same results as the original.

Written by the indexing model from the issue text.