This is the format used by ClickHouse, so it's useful to support this
too in addition to a valid JSON list of benchmark records.

### Testing

* Adding a test case
https://github.com/pytorch/test-infra/actions/runs/11927946564/job/33244134505#step:5:124
* I'm also testing this in
pytorch/pytorch#141087

A small fix I just realize after #141087.

Pull Request resolved: #141145
Approved by: https://github.com/malfet

Summary:
This uploads the MPS benchmark results to benchmark database.  The data can then be queried, for example:

```
select benchmark, model, metric from oss_ci_benchmark_v3 where head_sha = '99a133116fee15aa1467165f2b209b37da53f189' and metric.name in ['eager_peak_mem', 'dynamo_peak_mem', 'speedup'] and model.name = 'BERT_pytorch'
```

I'm documenting the JSON format at https://github.com/pytorch/pytorch/wiki/How-to-integrate-with-PyTorch-OSS-benchmark-database

### Testing

Locally,

```
PYTHONPATH=/Users/huydo/Storage/mine/benchmark python benchmarks/dynamo/torchbench.py --performance --only resnet152 --backend eager --training --devices mps --output test/test-reports/torchbench_training.csv
```

Workflow dispatch https://github.com/pytorch/pytorch/actions/runs/11927990520

X-link: pytorch/pytorch#141087
Approved by: https://github.com/malfet

Reviewed By: izaitsevfb

Differential Revision: D66262910

Pulled By: huydhn

fbshipit-source-id: 318fbb0ded0e85742aa441e023f5abf906398997

Summary:
A small fix I just realize after pytorch/pytorch#141087.

X-link: pytorch/pytorch#141145
Approved by: https://github.com/malfet

Reviewed By: izaitsevfb

Differential Revision: D66262943

Pulled By: huydhn

fbshipit-source-id: 4ccec00d23bac753fbba3ad1a39de1569efd11f8

)

Similar to #141087, this uploads the micro benchmark results to benchmark database with its new schema v3. The data can then be queried.

~I'm testing with `inductor-micro-benchmark-x86` which should be sufficient because `inductor-micro-benchmark` is broken atm.  The CSV output stays for now until the dashboard is migrated to schema v3.~ #141747 has been resolved, so inductor-micro-benchmark should work now

Pull Request resolved: #141745
Approved by: https://github.com/yanboliang

This uploads the MPS benchmark results to benchmark database.  The data can then be queried, for example:

```
select benchmark, model, metric from oss_ci_benchmark_v3 where head_sha = '99a133116fee15aa1467165f2b209b37da53f189' and metric.name in ['eager_peak_mem', 'dynamo_peak_mem', 'speedup'] and model.name = 'BERT_pytorch'
```

I'm documenting the JSON format at https://github.com/pytorch/pytorch/wiki/How-to-integrate-with-PyTorch-OSS-benchmark-database

### Testing

Locally,

```
PYTHONPATH=/Users/huydo/Storage/mine/benchmark python benchmarks/dynamo/torchbench.py --performance --only resnet152 --backend eager --training --devices mps --output test/test-reports/torchbench_training.csv
```

Workflow dispatch https://github.com/pytorch/pytorch/actions/runs/11927990520

Pull Request resolved: pytorch#141087
Approved by: https://github.com/malfet

…#141145)

A small fix I just realize after pytorch#141087.

Pull Request resolved: pytorch#141145
Approved by: https://github.com/malfet

…rch#141745)

Similar to pytorch#141087, this uploads the micro benchmark results to benchmark database with its new schema v3. The data can then be queried.

~I'm testing with `inductor-micro-benchmark-x86` which should be sufficient because `inductor-micro-benchmark` is broken atm.  The CSV output stays for now until the dashboard is migrated to schema v3.~ pytorch#141747 has been resolved, so inductor-micro-benchmark should work now

Pull Request resolved: pytorch#141745
Approved by: https://github.com/yanboliang