Google cloud session crashes when using cuML when clustering in fit

Nobody has claimed this yet.

I am using cuML as the hdbscan for clustering using gpu as I have 1 million short docs and I need to run with calculate_probabilities=True to get topic distributions.

My code is as follows:
```python
from google.colab import userdata
from cuml.cluster import HDBSCAN

api_key = 'openai-key'

hdbscan_model = HDBSCAN(min_samples=10, gen_min_span_tree=True, prediction_data=True)
docs = df['body']
topic_modeler = TopicModel()
topics = topic_modeler.train_topic_model(docs,
embeddings=embeddings,
min_topic_size=5,
nr_topics = None,
calculate_probabilities=True,
hdbscan_model=hdbscan_model,
representation_model='openAI',
api_key=api_key,
save_path=MODEL_PATH,
save_file= 'topic_model_min_topics_5_no_dc_subset')`
```

I see the following messages:

```
2025-05-22 12:32:05,122 - BERTopic - Dimensionality - Fitting the dimensionality reduction algorithm
2025-05-22 12:38:00,201 - BERTopic - Dimensionality - Completed ✓
2025-05-22 12:38:00,211 - BERTopic - Cluster - Start clustering the reduced embeddings
```

But a couple of minutes after starting clustering the kernel crashes without any messages.

Please find attached the logs. Any help would be appreciated.

[cuml_error.log](https://github.com/user-attachments/files/20392354/cuml_error.log)

Start with the provided Python snippet and cuml_error.log, then reproduce the BERTopic clustering stage with cuML HDBSCAN and calculate_probabilities=True in the reported Google Colab environment. Determine whether the crash is in BERTopic integration or the cuML/GPU runtime; done means documenting a confirmed cause and a scoped fix or workaround.

Written by the indexing model from the issue text.