[BUG] difference in onnx score when using batch

### SynapseML version

0.11.4

### System information

- **Language version** (e.g. python 3.8, scala 2.12): 3.10
- **Spark Version** (e.g. 3.2.3): 3.2.0
- **Spark Platform** (e.g. Synapse, Databricks): yarn

### Describe the problem

Hello, I would like to inquire if anyone has faced similar issues where scores are different when batch size is 1 and batch size eis higher number say 5000.
The score is consistently same for each batch size. So not a model issue. Score with batch size 1 matches that when produced with onnx runtime with pandas dataframe.

Difference in score between these two are also significant.

### Code to reproduce issue

Unfortunately will not be able to produce code.

### 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