Drop careless-gpu: the [cuda] pip extra install covers GPU and CPU nodes

The installed careless/0.5.4 venv already contains the NVIDIA wheels
([cuda] extra), which need only the node driver and fall back to CPU,
so a separate GPU module and cuda module dep are unnecessary. Revert
careless/build to careless[cuda] to match the actual install.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

careless-gpu/README.md

@@ -1,28 +0,0 @@
-# careless-gpu
-
-GPU build of careless (see `../careless/` for the CPU module and the full
-background). Scaling and merging of X-ray diffraction data with approximate
-Bayesian inference (variational merging), built on TensorFlow.
-
-- Upstream: https://github.com/rs-station/careless
-- License: MIT
-
-## How it differs from the careless module
-
-- CUDA is not pip-installed (no `[cuda]` extra, saves several GB): the
-  cluster `cuda/12.8.1` module provides it instead, declared as
-  `build_requires`/`runtime_deps` (same pattern as `jfjoch_viewer`).
-  `runtime_deps` matters because TensorFlow dlopens the CUDA libraries
-  when careless runs, so cuda must auto-load with this module.
-- If careless reports missing cuDNN at runtime, the cuda module does not
-  ship it; the fallback is installing `careless[cuda]` in `build` instead.
-
-The venv layout, uv usage, and cache handling are identical to `careless`.
-To share one download cache between both builds, run each with
-`UV_CACHE_DIR=<shared path> modbuild build <version>` — a cache outside
-PREFIX is kept, an in-PREFIX cache is removed after a successful install.
-
-## Adding a new version
-
-Add the version under `versions:` in `files/config.yaml`, then on a Ra
-login node: `modbuild build <version>` (modbuild/2.1.2).

careless-gpu/build

@@ -1,26 +0,0 @@
-#!/usr/bin/env modbuild
-
-pbuild::prep() {
-    :
-}
-
-pbuild::configure() {
-    # same uv pattern as the careless (CPU) module; see that build for the whys
-    UV="${UV:-$HOME/.local/bin/uv}"
-    export UV_CACHE_DIR="${UV_CACHE_DIR:-$PREFIX/.uv-cache}"
-    export UV_PYTHON_INSTALL_DIR="$PREFIX/python"
-    "$UV" venv --python 3.12 "$PREFIX/venv"
-}
-
-pbuild::compile() {
-    :
-}
-
-pbuild::install() {
-    UV="${UV:-$HOME/.local/bin/uv}"
-    export UV_CACHE_DIR="${UV_CACHE_DIR:-$PREFIX/.uv-cache}"
-    # no [cuda] extra: CUDA comes from the cuda/12.8.1 module
-    # (build_requires/runtime_deps in files/config.yaml), not from pip wheels
-    "$UV" pip install --python "$PREFIX/venv/bin/python" "careless==${V_PKG}"
-    [[ "$UV_CACHE_DIR" == "$PREFIX"* ]] && rm -rf "$UV_CACHE_DIR" || :
-}

careless-gpu/files/config.yaml

@@ -1,13 +0,0 @@
----
-format: 1
-careless-gpu:
-  defaults:
-    group: MX
-    overlay: base
-    relstage: unstable
-  versions:
-    0.5.4:
-      config:
-        relstage: unstable
-        build_requires: ["cuda/12.8.1"]
-        runtime_deps: ["cuda/12.8.1"]

careless-gpu/modulefile

@@ -1,16 +0,0 @@
-#%Module1.0
-
-module-whatis       "careless (GPU) - merging crystallography data with variational inference"
-module-url          "https://github.com/rs-station/careless"
-module-license      "MIT"
-module-maintainer   "MX Data <jiaxin.duan@psi.ch>"
-
-module-help     "
-careless: scaling and merging of X-ray diffraction data using
-approximate Bayesian inference (TensorFlow). GPU build: CUDA is
-provided by the cuda/12.8.1 module, auto-loaded as a runtime dep.
-Use on nodes with an NVIDIA GPU; for CPU nodes load careless instead.
-"
-
-setenv          CARELESS_ENV    $PREFIX/venv
-prepend-path    PATH            $PREFIX/venv/bin