Revert "docs(install): document cuTENSOR/cuQuantum as required for CUDA builds (#1130)"

[yingjerkao]

Reverts #1130 (merged as a0ddbbe2) at the maintainer's request.

docs/source/adv_install.rst returns to its pre-#1130 state: 71 lines removed, 9 restored — the exact inverse of the merge. No other file is touched, and nothing else on master depends on it.

Why this is a revert PR and not a reopen

#1130 was squash-merged, and GitHub cannot reopen a merged pull request:

X Pull request #1130 can't be reopened because it was already merged

So the merge is undone here instead. The original branch docs/1970-cuda-install-options has been restored at its original head 910e8261, so none of the work is lost and it can be carried into a fresh PR whenever the content is wanted again.

What #1130 had changed

For reference, so it can be reconstructed or reworked:

• reframed cuTENSOR/cuQuantum as required for CUDA builds rather than optional, documenting CUTENSOR_ROOT / CUQUANTUM_ROOT
• added a CUDA build-options table, plus a BACKEND_TORCH row
• corrected the default install prefix to ~/.local/cytnx (the page said /usr/local/cytnx)
• fixed a copy-paste error where the cuQuantum paragraph showed -DUSE_CUTENSOR=ON, and a "cuqauantum" typo
• noted that presets override these settings, and warned about [build] FindCUTENSOR/FindCUQUANTUM: exported CUTENSOR_ROOT/CUQUANTUM_ROOT silently overrides an explicit -D #1129

Refs #1130, #970.

[gemini-code-assist]

Caution

The consumer version of Gemini Code Assist on GitHub has been sunset. All code review activity has officially ceased.

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: be0ba6cee3

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[chatgpt-codex-connector]

Install the dependencies enabled by CUDA presets

A conda user who follows this GPU step and then uses the documented preset workflow cannot configure a CUDA build: openblas-cuda and mkl-cuda force both USE_CUTENSOR and USE_CUQUANTUM on in CMakePresets.json, while CMakeLists.txt aborts unless their packages and CUTENSOR_ROOT/CUQUANTUM_ROOT are available. Presenting cuTENSOR and cuQuantum as optional and installing only cuda therefore leaves the shipped CUDA presets unusable; restore the required packages and root-variable setup or explain how to disable those preset options.

Useful? React with 👍 / 👎.

[chatgpt-codex-connector]

Use the cuQuantum option in the cuQuantum instructions

For users configuring without a preset, this paragraph tells them to enable USE_CUTENSOR when they want cuQuantum. That option enables a different dependency and never activates cuQuantum functionality; the documented compile option here must be -DUSE_CUQUANTUM=ON.

Useful? React with 👍 / 👎.

[chatgpt-codex-connector]

Report the configured install prefix

The stated default disagrees with both CMakeLists.txt, which initializes CMAKE_INSTALL_PREFIX to ~/.local/cytnx, and the default preset in CMakePresets.json. Users who omit the option will therefore look under /usr/local/cytnx even though the project installs under their home directory.

Useful? React with 👍 / 👎.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 73.06%. Comparing base (cf539f0) to head (be0ba6c).
⚠️ Report is 1 commits behind head on master.
✅ All tests successful. No failed tests found.

Additional details and impacted files

@@           Coverage Diff           @@
##           master    #1134   +/-   ##
=======================================
  Coverage   73.06%   73.06%           
=======================================
  Files         224      224           
  Lines       27642    27642           
  Branches       71       71           
=======================================
  Hits        20196    20196           
  Misses       7425     7425           
  Partials       21       21           

Flag	Coverage Δ
cpp	73.18% <ø> (ø)
python	64.13% <ø> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

Components	Coverage Δ
C++ backend	72.54% <ø> (ø)
Python bindings	76.75% <ø> (ø)
Python package	64.13% <ø> (ø)

---

Continue to review full report in Codecov by Harness.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update cf539f0...be0ba6c. Read the comment docs.

🚀 New features to boost your workflow:

• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.