Add fast path for indices on non-invertible TransformedGrid

[marcosdanieldasilva]

This PR adds a major performance optimization for finding indices on a TransformedGrid, specifically when it has a non-invertible SequentialTransform (like a Rotate followed by a Morphological transform).

Previously, this situation would fall back to the generic findall(intersects(geometry), domain) method, which is extremely slow because it must perform a costly intersection check for every single cell.

This new, specialized method implements a much faster "partial revert" strategy. It introspects the transform sequence, finds the invertible component (using isinvertible), and reverts the input geometry using only that part. This "un-transforms" the geometry back to the original grid's coordinate space, allowing it to call the existing, highly-optimized indices(grid.mesh, ...) function on the simple, non-deformed grid. A new test is included to validate this exact scenario.

[juliohm]

Thank you @marcosdanieldasilva !

Added a few suggestions to generalize the code:

1. The transform can have multiple steps, so we need to combine all the revertible ones into a new transform in reverse order.
2. We can then just apply the created transform to the geometry.

We still need to check if any transform step needs special treatment.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 87.52%. Comparing base (7c2fbc3) to head (cfb2407).

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #1264      +/-   ##
==========================================
- Coverage   87.90%   87.52%   -0.38%     
==========================================
  Files         197      197              
  Lines        6240     6246       +6     
==========================================
- Hits         5485     5467      -18     
- Misses        755      779      +24     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[github-actions]

Benchmark Results (Julia vlts)

Time benchmarks

	master	897644d...	master / 897644d...
clipping/SutherlandHodgman	3.69 ± 0.18 μs	3.77 ± 1.5 μs	0.979 ± 0.39
discretization/simplexify	26.4 ± 2.5 ms	27.7 ± 2.3 ms	0.954 ± 0.12
intersection/ray-triangle	0.05 ± 0.01 μs	0.05 ± 0.01 μs	1 ± 0.28
sideof/ring/large	6.53 ± 0.031 μs	6.53 ± 0.01 μs	1 ± 0.005
sideof/ring/small	0.05 ± 0.001 μs	0.05 ± 0.001 μs	1 ± 0.028
topology/half-edge	2.73 ± 0.045 ms	2.72 ± 0.037 ms	1.01 ± 0.021
winding/mesh	0.0424 ± 0.0022 s	0.0429 ± 0.0021 s	0.988 ± 0.07
time_to_load	1.54 ± 0.022 s	1.52 ± 0.0076 s	1.01 ± 0.016

Memory benchmarks

	master	897644d...	master / 897644d...
clipping/SutherlandHodgman	0.053 k allocs: 4.97 kB	0.053 k allocs: 4.97 kB	1
discretization/simplexify	0.226 M allocs: 21.8 MB	0.226 M allocs: 21.8 MB	1
intersection/ray-triangle	0 allocs: 0 B	0 allocs: 0 B
sideof/ring/large	0 allocs: 0 B	0 allocs: 0 B
sideof/ring/small	0 allocs: 0 B	0 allocs: 0 B
topology/half-edge	18.1 k allocs: 2.92 MB	18.1 k allocs: 2.92 MB	1
winding/mesh	0.231 M allocs: 23 MB	0.231 M allocs: 23 MB	1
time_to_load	0.153 k allocs: 14.5 kB	0.153 k allocs: 14.5 kB	1

[github-actions]

Benchmark Results (Julia v1)

Time benchmarks

	master	897644d...	master / 897644d...
clipping/SutherlandHodgman	2.6 ± 0.25 μs	2.6 ± 0.29 μs	1 ± 0.15
discretization/simplexify	23.5 ± 3.8 ms	23.9 ± 4 ms	0.984 ± 0.23
intersection/ray-triangle	0.06 ± 0.009 μs	0.05 ± 0.001 μs	1.2 ± 0.18
sideof/ring/large	6.84 ± 0.011 μs	6.78 ± 0.021 μs	1.01 ± 0.0035
sideof/ring/small	0.06 ± 0 μs	0.061 ± 0.01 μs	0.984 ± 0.16
topology/half-edge	2.76 ± 0.18 ms	2.78 ± 0.17 ms	0.994 ± 0.091
winding/mesh	0.0405 ± 0.0029 s	0.0403 ± 0.0031 s	1 ± 0.1
time_to_load	1.48 ± 0.013 s	1.48 ± 0.011 s	1 ± 0.011

Memory benchmarks

	master	897644d...	master / 897644d...
clipping/SutherlandHodgman	0.064 k allocs: 5.55 kB	0.064 k allocs: 5.55 kB	1
discretization/simplexify	0.321 M allocs: 21.7 MB	0.321 M allocs: 21.7 MB	1
intersection/ray-triangle	0 allocs: 0 B	0 allocs: 0 B
sideof/ring/large	0 allocs: 0 B	0 allocs: 0 B
sideof/ring/small	0 allocs: 0 B	0 allocs: 0 B
topology/half-edge	25.9 k allocs: 3.17 MB	25.9 k allocs: 3.17 MB	1
winding/mesh	0.326 M allocs: 22.9 MB	0.326 M allocs: 22.9 MB	1
time_to_load	0.145 k allocs: 11 kB	0.145 k allocs: 11 kB	1