[SYCL] Add -vc-codegen compile option for ESIMD images

[DenisBakhvalov]

This change is a part of 3 patches:
Patch (1/3): [new property] #3072
Patch (2/3): [sycl-post-link] #3073
Patch (3/3): [DPCPP RT] this PR

Please review only patch f8fdd3c

[smaslov-intel]

LGTM. NOTE: the pi changes are already a part of #3072

[DenisBakhvalov]

NOTE: The current PR will not have any immediate impact on functionality since we guard splitting SYCL and ESIMD code in sycl-post-link tool under an option (-split-esimd). This PR will make it a default: #3086.

@kbobrovs, @romanovvlad, @bader, My current plan is to test it using SYCL_PI_TRACE output. I.e. put the functional test in the llvm-test-suite (see intel/llvm-test-suite#110). Is it OK for this patch (f8fdd3c) to go without the immediate test? Eventually, every SYCL/ESIMD test will be "de facto" testing this code.

[romanovvlad]

NOTE: The current PR will not have any immediate impact on functionality since we guard splitting SYCL and ESIMD code in sycl-post-link tool under an option (-split-esimd). This PR will make it a default: #3086.

@kbobrovs, @romanovvlad, @bader, My current plan is to test it using SYCL_PI_TRACE output. I.e. put the functional test in the llvm-test-suite (see intel/llvm-test-suite#110). Is it OK for this patch (f8fdd3c) to go without the immediate test? Eventually, every SYCL/ESIMD test will be "de facto" testing this code.

OK with me.

[kbobrovs]

Looks good to me. but the identifier might change if we rename SYCL/Is ESIMD Image category change.

[DenisBakhvalov]

Closing this PR since there were significant changes to #3072 (the PR that precedes this). I will submit a new PR once #3072 will get merged.