Missing CSE in TorchInductor

### 🐛 Describe the bug

A performance issue when using `torch.compile`. The compiled kernel introduces redundant computations that should be eliminated by common subexpression elimination (CSE).

Minimal reproducer:
```python
import argparse
import time
import torch
import torch.nn.functional as F
import torch.nn as nn

def compute_var_and_std(x: torch.Tensor) -> torch.Tensor:
var = torch.var(x, dim=-1)
std = torch.std(x, dim=-1)
return var + std

def benchmark(fn, *args, warmup=20, iters=100):
for _ in range(warmup):
fn(*args)
if torch.cuda.is_available():
torch.cuda.synchronize()
start_time = time.perf_counter()
for _ in range(iters):
fn(*args)
if torch.cuda.is_available():
torch.cuda.synchronize()
end_time = time.perf_counter()
return ((end_time - start_time) / iters) * 1000

def main() -> None:
parser = argparse.ArgumentParser()
parser.add_argument("--batch-size", type=int, default=64)
parser.add_argument("--vocab-size", type=int, default=32768)
parser.add_argument("--device", type=str, default="cuda" if torch.cuda.is_available() else "cpu")
parser.add_argument("--dtype", type=str, default="float16")
args = parser.parse_args()

device = torch.device(args.device)
dtype = torch.float16 if args.dtype == "float16" else torch.float32
B, V = args.batch_size, args.vocab_size

print(f"Device: {device} | Dtype: {dtype} | Batch: {B} | Vocab: {V}\n")

x = torch.randn(B, V, device=device, dtype=dtype)
x_pair = (torch.randn(B, V, device=device, dtype=dtype),
torch.randn(B, V, device=device, dtype=dtype))
labels = torch.randint(0, V, (B,), device=device)

cases = [
("var_std", compute_var_and_std, (x,)),
]

compile_modes = [
("Eager", None),
("Default", {}),
("ReduceOH", {"mode": "reduce-overhead"}),
("MaxAutotune", {"mode": "max-autotune"}),
]

for case_name, fn, fn_args in cases:
print(f"{'='*20} {case_name} {'='*20}")
eager_ms = None
for mode_name, compile_kwargs in compile_modes:
if compile_kwargs is None:
compiled_fn = fn
else:
compiled_fn = torch.compile(fn, **compile_kwargs)
try:
compiled_fn(*fn_args) # trigger compilation
ms = benchmark(compiled_fn, *fn_args)
if mode_name == "Eager":
eager_ms = ms
speedup = eager_ms / ms if eager_ms else float("nan")
print(f" {mode_name:12s}: {ms:8.4f} ms speedup={speedup:.3f}x")
except Exception as e:
print(f" {mode_name:12s}: FAILED -> {str(e)[:60]}")
print()

if __name__ == "__main__":
main()
```

Performance Rresult:
```
Device: cuda | Dtype: torch.float16 | Batch: 64 | Vocab: 32768

==================== var_std ====================
Eager : 0.0289 ms speedup=1.000x
Default : 0.0866 ms speedup=0.334x
ReduceOH : 0.0691 ms speedup=0.418x
MaxAutotune : 0.0688 ms speedup=0.420x
```

Inspecting the generated output_code.py, the fused Triton kernel (e.g., triton_red_fused_std_var_0) initializes two independent Welford state machines:
```python
tmp2_mean = tl.zeros([XBLOCK, RBLOCK], tl.float32)
tmp2_m2 = tl.zeros([XBLOCK, RBLOCK], tl.float32)

# Second set (for std)
tmp7_mean = tl.zeros([XBLOCK, RBLOCK], tl.float32)
tmp7_m2 = tl.zeros([XBLOCK, RBLOCK], tl.float32)
```
However, TorchInductor should detect that torch.var(x) and torch.std(x) share the same reduction statistics. The intermediate results can be computed once and reused.
By manually editing the generated Triton kernel, we prove the importance of CSE:
```
--- Validate Correctness ---
Correctness: Pass

--- Test Performance ---
Original Code Avg Latency: 0.0515 ms
Manual CSE Code Avg Latency: 0.0411 ms
Speedup: 20.24%
```

I noticed that PyTorch Inductor has a design intent for CSE, but it does not seem to be well-implemented here. Is this expected behavior, or a missing optimization in TorchInductor?

### Versions

```
PyTorch version: 2.10.0.dev20251021+cu128
Is debug build: False
CUDA used to build PyTorch: 12.8
ROCM used to build PyTorch: N/A

OS: Ubuntu 22.04.4 LTS (x86_64)
GCC version: (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0
Clang version: Could not collect
CMake version: version 3.22.1
Libc version: glibc-2.35

Python version: 3.10.19 (main, Oct 21 2025, 16:43:05) [GCC 11.2.0] (64-bit runtime)
Python platform: Linux-6.8.0-40-generic-x86_64-with-glibc2.35
Is CUDA available: True
CUDA runtime version: 11.8.89
CUDA_MODULE_LOADING set to:
GPU models and configuration:
GPU 0: NVIDIA RTX A6000
GPU 1: NVIDIA RTX A6000

Nvidia driver version: 570.211.01
cuDNN version: Could not collect
Is XPU available: False
HIP runtime version: N/A
MIOpen runtime version: N/A
Is XNNPACK available: True

CPU:
Architecture: x86_64
CPU op-mode(s): 32-bit, 64-bit
Address sizes: 46 bits physical, 48 bits virtual
Byte Order: Little Endian
CPU(s): 96
On-line CPU(s) list: 0-95
Vendor ID: GenuineIntel
Model name: Intel(R) Xeon(R) Gold 6248R CPU @ 3.00GHz
CPU family: 6
Model: 85
Thread(s) per core: 2
Core(s) per socket: 24
Socket(s): 2
Stepping: 7
BogoMIPS: 6000.00
Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid dca sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_fault epb cat_l3 cdp_l3 intel_ppin ssbd mba ibrs ibpb stibp ibrs_enhanced tpr_shadow flexpriority ept vpid ept_ad fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid cqm mpx rdt_a avx512f avx512dq rdseed adx smap clflushopt clwb intel_pt avx512cd avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local dtherm ida arat pln pts vnmi pku ospke avx512_vnni md_clear flush_l1d arch_capabilities
Virtualization: VT-x
L1d cache: 1.5 MiB (48 instances)
L1i cache: 1.5 MiB (48 instances)
L2 cache: 48 MiB (48 instances)
L3 cache: 71.5 MiB (2 instances)
NUMA node(s): 2
NUMA node0 CPU(s): 0,2,4,6,8,10,12,14,16,18,20,22,24,26,28,30,32,34,36,38,40,42,44,46,48,50,52,54,56,58,60,62,64,66,68,70,72,74,76,78,80,82,84,86,88,90,92,94
NUMA node1 CPU(s): 1,3,5,7,9,11,13,15,17,19,21,23,25,27,29,31,33,35,37,39,41,43,45,47,49,51,53,55,57,59,61,63,65,67,69,71,73,75,77,79,81,83,85,87,89,91,93,95
Vulnerability Gather data sampling: Mitigation; Microcode
Vulnerability Itlb multihit: KVM: Mitigation: VMX disabled
Vulnerability L1tf: Not affected
Vulnerability Mds: Not affected
Vulnerability Meltdown: Not affected
Vulnerability Mmio stale data: Mitigation; Clear CPU buffers; SMT vulnerable
Vulnerability Reg file data sampling: Not affected
Vulnerability Retbleed: Mitigation; Enhanced IBRS
Vulnerability Spec rstack overflow: Not affected
Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1: Mitigation; usercopy/swapgs barriers and __user pointer sanitization
Vulnerability Spectre v2: Mitigation; Enhanced / Automatic IBRS; IBPB conditional; RSB filling; PBRSB-eIBRS SW sequence; BHI SW loop, KVM SW loop
Vulnerability Srbds: Not affected
Vulnerability Tsx async abort: Mitigation; TSX disabled

Versions of relevant libraries:
[pip3] numpy==2.2.6
[pip3] nvidia-cublas-cu11==11.11.3.6
[pip3] nvidia-cublas-cu12==12.8.4.1
[pip3] nvidia-cuda-cupti-cu11==11.8.87
[pip3] nvidia-cuda-cupti-cu12==12.8.90
[pip3] nvidia-cuda-nvrtc-cu11==11.8.89
[pip3] nvidia-cuda-nvrtc-cu12==12.8.93
[pip3] nvidia-cuda-runtime-cu11==11.8.89
[pip3] nvidia-cuda-runtime-cu12==12.8.90
[pip3] nvidia-cudnn-cu11==9.1.0.70
[pip3] nvidia-cudnn-cu12==9.10.2.21
[pip3] nvidia-cufft-cu11==10.9.0.58
[pip3] nvidia-cufft-cu12==11.3.3.83
[pip3] nvidia-curand-cu11==10.3.0.86
[pip3] nvidia-curand-cu12==10.3.9.90
[pip3] nvidia-cusolver-cu11==11.4.1.48
[pip3] nvidia-cusolver-cu12==11.7.3.90
[pip3] nvidia-cusparse-cu11==11.7.5.86
[pip3] nvidia-cusparse-cu12==12.5.8.93
[pip3] nvidia-cusparselt-cu12==0.7.1
[pip3] nvidia-nccl-cu11==2.21.5
[pip3] nvidia-nccl-cu12==2.27.5
[pip3] nvidia-nvjitlink-cu12==12.8.93
[pip3] nvidia-nvtx-cu11==11.8.86
[pip3] nvidia-nvtx-cu12==12.8.90
[pip3] pytorch-triton==3.5.0+git7416ffcb
[pip3] torch==2.10.0.dev20251021+cu128
[pip3] torchaudio==2.10.0.dev20251021+cu128
[pip3] torchvision==0.25.0.dev20251021+cu128
[pip3] triton==3.0.0
[conda] numpy 2.2.6 pypi_0 pypi
[conda] nvidia-cublas-cu11 11.11.3.6 pypi_0 pypi
[conda] nvidia-cublas-cu12 12.8.4.1 pypi_0 pypi
[conda] nvidia-cuda-cupti-cu11 11.8.87 pypi_0 pypi
[conda] nvidia-cuda-cupti-cu12 12.8.90 pypi_0 pypi
[conda] nvidia-cuda-nvrtc-cu11 11.8.89 pypi_0 pypi
[conda] nvidia-cuda-nvrtc-cu12 12.8.93 pypi_0 pypi
[conda] nvidia-cuda-runtime-cu11 11.8.89 pypi_0 pypi
[conda] nvidia-cuda-runtime-cu12 12.8.90 pypi_0 pypi
[conda] nvidia-cudnn-cu11 9.1.0.70 pypi_0 pypi
[conda] nvidia-cudnn-cu12 9.10.2.21 pypi_0 pypi
[conda] nvidia-cufft-cu11 10.9.0.58 pypi_0 pypi
[conda] nvidia-cufft-cu12 11.3.3.83 pypi_0 pypi
[conda] nvidia-curand-cu11 10.3.0.86 pypi_0 pypi
[conda] nvidia-curand-cu12 10.3.9.90 pypi_0 pypi
[conda] nvidia-cusolver-cu11 11.4.1.48 pypi_0 pypi
[conda] nvidia-cusolver-cu12 11.7.3.90 pypi_0 pypi
[conda] nvidia-cusparse-cu11 11.7.5.86 pypi_0 pypi
[conda] nvidia-cusparse-cu12 12.5.8.93 pypi_0 pypi
[conda] nvidia-cusparselt-cu12 0.7.1 pypi_0 pypi
[conda] nvidia-nccl-cu11 2.21.5 pypi_0 pypi
[conda] nvidia-nccl-cu12 2.27.5 pypi_0 pypi
[conda] nvidia-nvjitlink-cu12 12.8.93 pypi_0 pypi
[conda] nvidia-nvtx-cu11 11.8.86 pypi_0 pypi
[conda] nvidia-nvtx-cu12 12.8.90 pypi_0 pypi
[conda] pytorch-triton 3.5.0+git7416ffcb pypi_0 pypi
[conda] torch 2.10.0.dev20251021+cu128 pypi_0 pypi
[conda] torchaudio 2.10.0.dev20251021+cu128 pypi_0 pypi
[conda] torchvision 0.25.0.dev20251021+cu128 pypi_0 pypi
```

cc @jerryzh168 @chauhang @penguinwu @voznesenskym @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @ipiszy @kadeng @muchulee8 @amjames @aakhundov @coconutruben @jataylo