Add support for Python3.14, drop support for 3.9 and 3.10

[tsmbland]

Description

Updates dependencies to use the latest versions of pandas, xarray, numpy and scipy. As a result, I've had to drop support for Python 3.9 and 3.10 as these are not supported by the latest package versions. There are also a few syntax changes related to pandas and xarray (in particular, xarray was warning about upcoming changes to some default parameters, so I've set these explicitly to avoid unwanted behaviour changes).

As shown by pandas 3.0, it's likely that any further major version updates to these packages will break things, so I've capped the upper versions to prevent unwanted breakages (e.g. when pandas 4 comes out). Not sure why I wasn't doing this already, to be honest.

Since this is a major breaking change, I'll wait for the next major release (i.e. v1.6.0) to merge this, whenever the time comes.

Fixes #825

Type of change

• New feature (non-breaking change which adds functionality)
• Optimization (non-breaking, back-end change that speeds up the code)
• Bug fix (non-breaking change which fixes an issue)
• Breaking change (whatever its nature)

Key checklist

• All tests pass: $ python -m pytest
• The documentation builds and looks OK: $ python -m sphinx -b html docs docs/build

Further checks

• Code is commented, particularly in hard-to-understand areas
• Tests added that prove fix is effective or that feature works

[Copilot]

Pull request overview

This PR updates MUSE’s supported Python/runtime stack to accommodate newer versions of core scientific dependencies (notably pandas/xarray), adding Python 3.14 support while dropping Python 3.9/3.10. It also adjusts xarray concatenation/merge behavior and refreshes CI/docs to match the new support window.

Changes:

• Update Python support metadata (pyproject classifiers / requires-python) and refresh CI + ReadTheDocs + installation docs to Python 3.11–3.14.
• Update xarray concat(...)/merge(...) calls to set explicit join/compat/coords/data_vars behavior (and adjust tests accordingly).
• Minor test/data-selection updates (e.g., pandas .isin(...) usage) and warning filtering cleanup.

Reviewed changes

Copilot reviewed 15 out of 15 changed files in this pull request and generated 6 comments.

Show a summary per file

File	Description
pyproject.toml	Raises minimum Python to 3.11, adds 3.14 classifier, updates dependency bounds.
src/muse/utilities.py	Adds explicit xr.concat parameters to stabilize concat behavior under newer xarray.
src/muse/filters.py	Refactors initializer asset indexing logic (potentially problematic).
src/muse/readers/csv.py	Uses join="outer" for merges to handle newer xarray merge defaults.
src/muse/readers/toml.py	Uses join="outer" when merging trade data into technodata.
tests/test_timeslice_output.py	Adjusts filtering to use .isin(...) for technology selection.
tests/test_demand_share.py	Adds explicit xr.concat parameters.
tests/conftest.py	Removes a Python 3.9-specific warning ignore.
docs/installation/virtual-env-based.rst	Updates example Python version and minor text edit.
docs/installation/pipx-based.rst	Updates stated compatible Python versions and pyenv examples.
.readthedocs.yaml	Bumps docs build Python to 3.11.
.github/workflows/ci.yml	Updates test matrix to 3.11/3.14 and Codecov gating.
.github/workflows/regression_tests.yml	Updates regression test matrix to 3.11/3.14.
.github/workflows/publish.yml	Updates standalone build Python to 3.11.
.github/workflows/documentation.yml	Bumps docs workflow Python to 3.14.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[tsmbland]

Not sure. This is functionally equivalent to before so not introducing any new bugs at least. This function doesn't seem to be used anyway in any of the examples, so may be able to delete

[Copilot]

Pull request overview

Copilot reviewed 49 out of 49 changed files in this pull request and generated 1 comment.

Comments suppressed due to low confidence (1)

src/muse/utilities.py:440

• nametuple_to_dict now accepts tuple, but the fallback branch calls nametup.copy() and then iterates .items(), which will raise at runtime for a plain tuple (and any non-Mapping without _asdict). Consider narrowing the annotation back to a Mapping/NamedTuple-like type (e.g., a protocol exposing _asdict) or making the fallback explicitly validate isinstance(nametup, Mapping) and raise a clear TypeError otherwise.