Improved optional dependencies handling

[probberechts]

This PR improves how optional dependencies are handled in kloppy.

Previously, checking for optional packages required scattering try/except ImportError blocks throughout the codebase. This PR introduces a centralized utility, import_optional_dependency, which elegantly handles missing packages by:

1. Checking if the package is installed.
2. Verifying that the installed version meets kloppy's minimum version requirements. (I'm not sure what they are currently, but these can be filled in later)
3. Raising a clear error message telling the user how to install the missing package (e.g., "Use pip
   or conda to install pandas").

Code Example:

  from kloppy._utils.optional import import_optional_dependency

  # This will either return the module or raise a helpful ImportError
  pd = import_optional_dependency("pandas")

  # You can also customize the error message or behavior
  nx = import_optional_dependency(
      "networkx",
      extra="This is required for regexp pattern matching.",
      on_version="warn" # Warns instead of raising if the version is too old
  )

It also adds a @skip_if_no testing decorator to easily skip unit tests in environments where these optional dependencies aren't installed. In the future, this will allow adding a dedicated CI job that installs only the core kloppy dependencies and tests that the base library functions perfectly for users who install the minimal version.