UPSTREAM PR #17510: Build llama.cpp using MSVC - Visual Studio 17 2022

[loci-dev]

Mirrored from ggml-org/llama.cpp#17510

Added detailed instructions for building libcurl and ccache dependencies.

Cheers,

[loci-agentic-ai]

Explore the complete analysis inside the Version Insights

Performance Analysis Summary - PR #329

Condition 1 Applied: No Performance Impact

This PR adds Windows MSVC build documentation only. No source code, build configuration, or performance-critical functions were modified. Zero impact on inference performance, tokens per second, or power consumption metrics.

[loci-agentic-ai]

Explore the complete analysis inside the Version Insights

Performance Analysis Summary - PR #329

Condition 1 Applied: No Performance Impact

This PR adds Windows MSVC build documentation only. No source code, build configuration, or performance-critical functions were modified. Zero impact on inference performance, tokens per second, or power consumption metrics.

[loci-agentic-ai]

Explore the complete analysis inside the Version Insights

Performance Analysis Summary: PR #329

Condition 1 Applied (No performance impact detected)

This PR adds MSVC build documentation only. No source code modifications were made to any performance-critical functions. Analysis confirms 0% change across all performance metrics: response time, throughput time, and power consumption remain identical between versions. All 16 binaries show stable energy profiles. No impact on inference performance or tokens per second.