[BUG]: CUDA OOM error after data extraction from certain number of articles using extract_composition_property_data function

[aritraroy24]

Describe the bug
CUDA out-of-memory (OOM) error during the data extraction processing using the extract_composition_property_data function.

To Reproduce
Steps to reproduce the behavior:

1. The previous steps [metadata collection and article processing] should be done beforehand to have the extracted data saved in a CSV file containing the articles with the required property.
2. Run the extract_composition_property_data function for multiple articles.
3. After ~45-50 articles, it gives the following error:

ERROR - Error in MaterialsFlow processing for DOI: 10.XXXX/XXXXXXXX. CUDA error: out of memory
Search for `cudaErrorMemoryAllocation' in https://docs.nvidia.com/cuda/cuda-runtime-api/group__CUDART__TYPES.html for more information.
CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect.
For debugging consider passing CUDA_LAUNCH_BLOCKING=1
Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions

Expected behavior
The script should not give the OOM error, and all the articles should be processed without failing.

System & Environment

• OS: Windows 11
• Processor: 13th Gen Intel(R) Core(TM) i9-13900HX (2.20 GHz)
• Installed RAM: 32 GB
• GPU Enabled: No
• Terminal: PowerShell
• Python environment: 3.12.10 (through Python virtual environment)

Additional context
Issue #1 may be related to this issue.