[llvm] [offload][SYCL] Add SYCL Module splitting. (PR #131347)

[Maksim Sabianin via llvm-commits]

maksimsab wrote:

Hi @Pierre-vh 

Currently, in the downstream, our SYCL extension is capable of being compiled into SPIR-V, Intel CPU/GPU, AMDGPU, and NVPTX. We are planning to upstream support for all of these. The one missing feature of this patch, which will be added later, is the splitting by optional kernel features ([spec](https://registry.khronos.org/SYCL/specs/sycl-2020/html/sycl-2020.html#sec:optional-kernel-features)). I think that justifies my choice of the component. Anyway, I am interested in your opinion.

The possible use case:
1) User writes 2  kernels:
* kernel A, which uses fp16
* kernel B, which doesn't use fp16.
2) If the user wants to JIT compile kernel B for a device that doesn't support fp16, then it is required that these kernels be separated from each other."

The separation is being done in the Module splitting algorithm.

https://github.com/llvm/llvm-project/pull/131347