feat(test): add support for TAP (Test Anything Protocol)

[jenstroeger]

The interesting thing about TAP (Test Anything Protocol) is that test output from different sources can be consolidated and analyzed as one. This change adds a pytest plugin which generated TAP output for tests, thus enabling TAP compatible test output.

However, I think before we would consider this change we need to track PR python-tap/pytest-tap#91 (and maybe PR python-tap/pytest-tap#92).

[jenstroeger]

With pytest-tap v3.5 released we can now capture and dump the output of individual tests as well:

> make test
Run unit tests...........................................................Passed
- hook id: pytest
- duration: 0.84s

TAP version 13
1..3
ok 1 src/package/something.py::[doctest] package.something.Something.do_something
# --- Captured Log ---
# 
# --- Captured Out ---
# 
# --- Captured Err ---
#
ok 2 tests/test_something.py::test_something
# --- Captured Log ---
# 
# --- Captured Out ---
# 
# --- Captured Err ---
#
ok 3 docs/source/index.rst::[doctest] index.rst
# --- Captured Log ---
# 
# --- Captured Out ---
# 
# --- Captured Err ---
#

However, some pytest plugins also print stats after the tests have been run:

---------- coverage: platform darwin, python 3.13.1-final-0 ----------
Name                       Stmts   Miss Branch BrPart  Cover   Missing
----------------------------------------------------------------------
src/package/__init__.py        1      0      0      0   100%
src/package/something.py       4      0      0      0   100%
----------------------------------------------------------------------
TOTAL                          5      0      0      0   100%

Required test coverage of 100.0% reached. Total coverage: 100.00%
============================ Hypothesis Statistics =============================

tests/test_something.py::test_something:

  - during generate phase (0.00 seconds):
    - Typical runtimes: ~ 0-1 ms, of which < 1ms in data generation
    - 2 passing examples, 0 failing examples, 0 invalid examples

  - Stopped because nothing left to do


============================== slowest durations ===============================
0.01s call     tests/test_something.py::test_something
0.00s call     src/package/something.py::package.something.Something.do_something
0.00s setup    src/package/something.py::package.something.Something.do_something
0.00s call     docs/source/index.rst::index.rst
0.00s teardown src/package/something.py::package.something.Something.do_something
0.00s setup    docs/source/index.rst::index.rst
0.00s teardown docs/source/index.rst::index.rst
0.00s setup    tests/test_something.py::test_something
0.00s teardown tests/test_something.py::test_something
============================== 3 passed in 0.08s ===============================

and those are still missing. We’ll have to dig around some more and check if we can capture those as well, and output them…