[Group Partitioner] Optimize Speed

[mcr229]

We do some optimizations on the group partitioner to improve its speed. Theoretically, since we've pre-grouped some partitions already, we should be faster than the capability based partitioner. For example, if we a dynamically quantized group, this could contain

[ choose_q_params, quantize, dequantize, op, dequantize_weight, weight, bias, get_item_scale, get_item_zp]

9 nodes. In capability-based partitioner they will have to run DFS on all 9 nodes in order to group these together. Based on the hints and purpose of the group based partitioenr, we don't perform these checks and instead group all these 9 nodes, saving time by avoiding these checks. Some stats when partitioning the mobile bert model:

elpased time old partitioner: 65.3421
old_partitioner num partitions: 170


elpased time new partitioner: 5.1964
new_partitioner num partitions: 170

we see a 13x improvement in partitioning when using the group based partitioner, while still partitioning around the same number of nodes.

Differential Revision: D79020720

[mcr229]

Stack from ghstack (oldest at bottom):

• [Group Partitioner] leverage group partitioner for config-based partitioner #12845
• -> [Group Partitioner] Optimize Speed #12844

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/12844

• 📄 Preview Python docs built from this PR

Note: Links to docs will display an error until the docs builds have been completed.

✅ No Failures

As of commit ec7a6b5 with merge base 21c8e67 ():
💚 Looks good so far! There are no failures yet. 💚

This comment was automatically generated by Dr. CI and updates every 15 minutes.

[github-actions]

This PR needs a release notes: label

If your change should be included in the release notes (i.e. would users of this library care about this change?), please use a label starting with release notes:. This helps us keep track and include your important work in the next release notes.

To add a label, you can comment to pytorchbot, for example
@pytorchbot label "release notes: none"

For more information, see
https://github.com/pytorch/pytorch/wiki/PyTorch-AutoLabel-Bot#why-categorize-for-release-notes-and-how-does-it-work.

[mcr229]

@mcr229 has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator.