[MNT] Move pytest configuration to pyproject.toml (#1851)

### Description

Consolidated pytest configuration into pyproject.toml for better
structure and to comply with modern Python tooling practices. Removed
pytest.ini as it is no longer needed after the migration.

Author: Borda

pyproject.toml

@@ -210,3 +210,31 @@ ignore_missing_imports = true
 no_implicit_optional = true
 check_untyped_defs = true
 cache_dir = ".cache/mypy/"
+
+[tool.pytest.ini_options]
+addopts = [
+    "-rsxX",
+    "-vv",
+    "--cov-config=.coveragerc",
+    "--cov=pytorch_forecasting",
+    "--cov-report=html",
+    "--cov-report=term-missing:skip-covered",
+    "--no-cov-on-fail"
+]
+markers = []
+testpaths = ["tests/", "pytorch_forecasting/tests/"]
+log_cli_level = "ERROR"
+log_format = "%(asctime)s %(levelname)s %(message)s"
+log_date_format = "%Y-%m-%d %H:%M:%S"
+cache_dir = ".cache"
+filterwarnings = [
+    "ignore:Found \\d+ unknown classes which were set to NaN:UserWarning",
+    "ignore:Less than \\d+ samples available for \\d+ prediction times. Use ba:UserWarning",
+    "ignore:scale is below 1e-7 - consider not centering the data or using data with:UserWarning",
+    "ignore:You defined a `validation_step` but have no `val_dataloader`:UserWarning",
+    "ignore:ReduceLROnPlateau conditioned on metric:RuntimeWarning",
+    "ignore:The number of training samples \\(\\d+\\) is smaller than the logging interval Trainer\\(:UserWarning",
+    "ignore:The dataloader, [\\_\\s]+ \\d+, does not have many workers which may be a bottleneck.:UserWarning",
+    "ignore:Consider increasing the value of the `num_workers` argument`:UserWarning",
+    "ignore::UserWarning"
+]