[Lane 5 · docs] TransposedConv2d — operator doc partial for convTranspose2d #1377

Sub-issue of #1371 ([Feature]: TransposedConv2d — convTranspose2d op, CPU kernel, backward and StableHLO lowering).

Lane: 5 · Docs/DARC
Skill needed: AsciiDoc and enough math to explain a transposed convolution. No Kotlin needed.
Size: s (a few hours)
Blocked by: nothing

What to do

1. Create docs/modules/ROOT/partials/ops/tensorops/convtranspose2d.adoc with the math, intuition, examples and references tags. Copy the layout of docs/modules/ROOT/partials/ops/tensorops/matmul.adoc.
2. Explain the output-size formula, why it is not a true inverse of conv2d, and the checkerboard-artifact caveat; cite the PyTorch docs and Dumoulin & Visin, A guide to convolution arithmetic for deep learning (arXiv:1603.07285).
3. Regenerate the docs with ./gradlew :skainet-compile:skainet-compile-hlo:jvmTest --tests '*HloCoverageDumpTest' :skainet-compile:skainet-compile-dag:jvmTest --tests '*BackwardCoverageDumpTest' generateDocs, check the badge on the generated tensorops.adoc changes from ✖ to ⚠, and commit the changed docs/modules/ROOT/pages/reference/ files (CI fails if they are stale).

Acceptance

• Partial exists with all four tags filled, no placeholders
• Every citation link resolves
• Result reported back on the parent issue

Notes

DARC validation (✅) is set later by a reviewer who is not you — see the DARC workflow page.

Coverage matrix: docs/modules/ROOT/pages/reference/ops-status-matrix.adoc · How lanes work: Issue taxonomy