[benchmark] run flash MLA bench on H100

```python
import modal

app = modal.App("ds-flash-mla")

# We also define the dependencies for our Function by specifying an
# [Image](https://modal.com/docs/guide/images).

ds_flash_mla_image = (
# https://hub.docker.com/r/pytorch/pytorch/tags
modal.Image.from_registry("pytorch/pytorch:2.6.0-cuda12.6-cudnn9-devel", add_python="3.11")
.apt_install("git")
.run_commands(
"git clone https://github.com/deepseek-ai/FlashMLA.git",
"cd FlashMLA && python setup.py install",
)
.pip_install()
)

BENCH_DIR = "/data/bench"
bench_dir = modal.Volume.from_name("bench", create_if_missing=True)

# modal run src/bench/run_flash_mla.py::compute_cap
@app.function(
gpu="H100",
max_inputs=1, # new container each input, so we re-roll the GPU dice every time
)
async def compute_cap():
import subprocess

gpu = subprocess.run(
["nvidia-smi", "--query-gpu=name,compute_cap,memory.total", "--format=csv,noheader"],
check=True,
text=True,
stdout=subprocess.PIPE,
).stdout.strip()
print(gpu)
return gpu

@app.function(
gpu="H100",
retries=0,
image=ds_flash_mla_image,
volumes={BENCH_DIR: bench_dir},
timeout=1200, # default 300s
container_idle_timeout=1200,
)
def flash_mla(bench_name="ds_flash_mla_bench") -> str:
import subprocess
import os

cmd = "python tests/test_flash_mla.py".split(" ")
result = subprocess.run(cmd, capture_output=True, text=True, cwd="/FlashMLA")
print(result)
with open(os.path.join(BENCH_DIR, f"{bench_name}.log"),"w") as f:
f.write(result.stdout)

# modal run src/bench/run_flash_mla.py
@app.local_entrypoint()
def main(bench_name="ds_flash_mla_bench"):
flash_mla.remote(bench_name)
```
bench result: https://github.com/ai-bot-pro/achatbot/pull/123
gpu u know: https://modal.com/gpu-glossary/readme