[FEATURE SUPPORT] add gemm pytorch kernel

[LoserCheems]

Summary

• Introduces a GEMM helper function in PyTorch to perform matrix multiplication with scaling factors, enhancing the existing functionality.

Root Cause

• The need for a reusable GEMM implementation in PyTorch flows prompted the addition of this utility.

Changes

• Added a new gemm function in pytorch_ops/gemm.py that updates a tensor using matrix multiplication with alpha and beta scaling. Updated the README to reflect the implementation status.

Reproduction

• Not applicable as this is a new feature implementation.

Tests

• No new tests were specified; existing tests for related operations can be referenced.

Compatibility

• No migration concerns or backwards compatibility issues noted.

Checklist

• Linked issue provided [FEATURE REQUEST] gemm PyTorch kernel implementation #31
• Adds or updates tests
• Updates docs if needed
• No perf regressions