[KED-2212] Improve graph performance

[bru5]

Description

Improves performance of the graph layout engine.

Development notes

In my tests on a large graph (860 nodes, 1500 edges), reduces its render time by about 60% / 2.5x boost.

This refactoring improves graph performance by:

• using static base constraints and constants
• using fully defined constraint values rather than defaults to avoid run time cost
• reducing assignments for creating constraint objects
• reducing memory usage for constraint objects
• removing spread operator when creating constraints (it compiles to a slower polyfill)

This is a refactoring only, the algorithm and the output remains the same.

Another benefit is that these changes also generally improve clarity being more explicit and with more abstraction too.

QA notes

No change to behaviour or results other than performance.

Test using a very large pipeline and compare render time.

Checklist

• Read the contributing guidelines
• Opened this PR as a 'Draft Pull Request' if it is work-in-progress
• Updated the documentation to reflect the code changes
• Added new entries to the RELEASE.md file
• Added tests to cover my changes

Legal notice

• I acknowledge and agree that, by checking this box and clicking "Submit Pull Request":

• I submit this contribution under the Apache 2.0 license and represent that I am entitled to do so on behalf of myself, my employer, or relevant third parties, as applicable.

• I certify that (a) this contribution is my original creation and / or (b) to the extent it is not my original creation, I am authorised to submit this contribution on behalf of the original creator(s) or their licensees.

• I certify that the use of this contribution as authorised by the Apache 2.0 license does not violate the intellectual property rights of anyone else.

[studioswong]

It would be really helpful to add a comment to explain what each field under the Constraint object is used for and which parts of the graph it affects - would definitely be helpful to provide clarity for future development work.

[bru5]

Ah yes the docs for these fields are on the solve function in solver.js, I'll add a comment in this file to make note of that ten

[studioswong]

interesting work @bru5 - as someone new to this codebase, it would be really helpful if you could provide some context to the graph layout engine setup in the PR description, particularly the reasoning behind some of the refactoring work done here that results in the improvement of the graph performance - is this based on any algorithms / graphing theory? would definitely love to understand and explore the ideas behind the current graph setup.

Wondering if we could also add in some tests to test the expected constraint values based on the animals dataset? The test case setup would also help to provide clarity on the expected behaviour of the graph layout engine as well.

All in all, very interesting stuff!

[bru5]

@studioswong thanks for looking over the PR

Yeah I'd love to discuss the approach in more detail but I think discussion of the algorithm is better left out of this particular PR as it's purely a code refactoring of the existing algorithm.

The only changes made are already in the PR description, but that's a good point I've added a note to clarify that there is no algorithmic change.

See graph.test.js for the unit tests, which includes the constraint and solver system which the layout algorithm is built on.

Testing against a specific dataset like animals would require a sophisticated verification / evaluation algorithm which is not currently implemented because it is not straightforward, so evaluating against datasets is manual for now but this would be a nice future addition.

[richardwestenra]

Can confirm, significantly faster. Graph shape looks about the same. Spectacular work!

[richardwestenra]

Note: JS tests are failing.

Also I recommend merging main as the Python tests on that branch have been fixed.