docs: generate 1.12.0rc1 release notes (#3888)

Author: flying-sheep

ci/scripts/towncrier_automation.py

@@ -71,13 +71,15 @@ def main(argv: Sequence[str] | None = None) -> None:
         text=True,
         check=True,
     ).stdout.strip()
-    pr_description = "" if base_branch == "main" else "@meeseeksdev backport to main"
+    pr_description = "- [x] Release notes not necessary because: compiles release notes"
+    if base_branch != "main":
+        pr_description += "\n\n@meeseeksdev backport to main"
     branch_name = f"release_notes_{args.version}"
 
     # Create a new branch + commit
     subprocess.run(["git", "switch", "-c", branch_name], check=True)
     subprocess.run(["git", "add", "docs/release-notes"], check=True)
-    pr_title = f"(chore): generate {args.version} release notes"
+    pr_title = f"docs: generate {args.version} release notes"
     subprocess.run(["git", "commit", "-m", pr_title], check=True)
 
     # push

docs/release-notes/1.12.0rc1.md

@@ -0,0 +1,39 @@
+(v1.12.0rc1)=
+### 1.12.0rc1 {small}`2025-11-10`
+
+#### Breaking changes
+
+- Adopt the Scientific Python [deprecation schedule](https://scientific-python.org/specs/spec-0000/):
+  remove Python ≤3.11 support and officially add Python 3.13 and 3.14 support, require anndata≥0.10 {smaller}`P Angerer` ({pr}`3485`, {pr}`3874`)
+
+#### Features
+
+- Added `n_components` parameter to {func}`~scanpy.tl.tsne` {smaller}`Kitsune` ({pr}`2803`)
+- {func}`scanpy.pp.highly_variable_genes` flavors `seurat_v3` and `seurat_v3_paper` are now `dask`-compatible {smaller}`I Gold` ({pr}`3340`)
+- Add {mod}`zarr` support and `convert_strings_to_categoricals` parameter to {func}`scanpy.write` {smaller}`P Angerer` ({pr}`3498`)
+- Add support for {class}`scipy.sparse.csr_array` and {class}`scipy.sparse.csc_array` {smaller}`P Angerer` ({pr}`3563`)
+- Added a new `compressed` parameter to {func}`~scanpy.read_10x_mtx` to support reading uncompressed matrix files produced by tools like STARsolo {smaller}`T Kuthuru` ({pr}`3564`)
+- Make {func}`scanpy.get.aggregate` {doc}`dask:index` compatible with all aggregations except median {smaller}`I Gold` ({pr}`3700`)
+- Add `filter_unexpressed_genes` parameter to {func}`scanpy.pp.highly_variable_genes` {smaller}`M Mueller` ({pr}`3767`)
+- Add column `"n_obs_aggregated"` to {func}`scanpy.get.aggregate` output to show the total number of observations aggregated per group {smaller}`R Disa` ({pr}`3824`)
+- Add `method='jaccard'` for generating connectivities in {func}`scanpy.pp.neighbors` {smaller}`J Pintar` ({pr}`3831`)
+- Add in `csc`-in-{doc}`dask:index` support for {func}`scanpy.get.aggregate` {smaller}`I Gold` ({pr}`3872`)
+- Raised warnings now always point at user code, not internal scanpy code {smaller}`P Angerer` ({pr}`3876`)
+
+#### Miscellaneous changes
+
+- Replaced several internal utilities with their {mod}`fast_array_utils` counterparts {smaller}`P Angerer` ({pr}`3598`)
+- Deprecate {func}`scanpy.tl.louvain` {smaller}`P Angerer` ({pr}`3658`)
+- Remove support for the experimental and unmaintained `zappy` library {smaller}`P Angerer` ({pr}`3884`)
+
+#### Documentation
+
+- Update tutorial {doc}`/tutorials/experimental/dask` for {mod}`anndata` 0.12 (see {pr}`scverse/scanpy-tutorials#186`) {smaller}`I Gold` ({pr}`3675`)
+- Add a warning message to {func}`~scanpy.tl.rank_genes_groups` that it is not appropriate for most use-cases {smaller}`G Sturm` ({pr}`3792`)
+
+#### Bug fixes
+
+- Raise fewer redundant warnings, mainly in {mod}`scanpy.pl` and {mod}`scanpy.datasets` functions {smaller}`P Angerer` ({pr}`3724`)
+- Fix `swap_axes` argument used with `standard_scale` in {func}`scanpy.pl.dotplot` {smaller}`I Gold` ({pr}`3851`)
+- Allow operating on {attr}`~anndata.AnnData.obsm` arrays in {func}`scanpy.pp.normalize_total` and {func}`scanpy.pp.pca` {smaller}`P Angerer` ({pr}`3863`)
+- Fix {func}`~scanpy.pl.dpt_groups_pseudotime` {smaller}`P Angerer` ({pr}`3864`)