[RFC][CodeGen][CUDA]: Gate fast math intrinsic lowering behind target option

[ConvolutedDog]

Fix CUDA lowering of standard TIR math intrinsics so they use precise CUDA math functions by default instead of fast-math __*f functions. This fixes the default behavior reported in #19546, where operators such as tirx.exp could lower to __expf even though fast math was not explicitly requested.

This change adds a CUDA target attribute, enable_fast_math, which defaults to false. When the attribute is unset or false, standard math intrinsics lower through the normal CUDA math rule, for example expf, logf, sinf, cosf, powf, and rsqrtf for float32. When users explicitly enable the attribute on the target, the lowering pass also checks the cuda.fastmath.FLowerIntrinsic rules before the normal CUDA lowering rules.

Users can opt in to fast math by constructing a CUDA target with the attribute:

tvm.target.Target({"kind": "cuda", "enable_fast_math": True})
target = tvm.target.Target({
    "tag": "nvidia/nvidia-a100",
    "enable_fast_math": True,
})

The fast-math lowering path currently covers the CUDA math operators registered with cuda.fastmath.FLowerIntrinsic: tirx.exp, tirx.exp10, tirx.log, tirx.log2, tirx.log10, tirx.tan, tirx.cos, tirx.sin, tirx.tanh, and tirx.pow.

tirx.rsqrt is also registered for CUDA lowering so it maps to the CUDA reciprocal-square-root intrinsic instead of being legalized as 1 / sqrt(x).

Add CUDA codegen tests tests/python/codegen/test_target_codegen_cuda_fastmath.py that check the lowered IR, generated CUDA source, and runtime results for the supported math intrinsics across floating point dtypes and both default and fast-math targets.

[gemini-code-assist]

Code Review

This pull request introduces support for fast math intrinsics in the CUDA backend by adding an enable_fast_math attribute to the CUDA target and updating the intrinsic lowering pass to prioritize these variants. It includes registrations for several math operations such as exp, log, and sin, along with a new test suite. The review feedback correctly identifies a contradictory assertion in the target tests where an attribute was explicitly set to True but asserted as False, and it suggests fixing a recurring typo (num_inputes to num_inputs) throughout the new test file.

[tlopex]

cc @LeiWang1999