Update MHKiT pip install to install particular modules

[jmcvey3]

Describe the bug:

Firstly, MHKiT installer needs to be updated to use pyproject.toml since the setup.py installation process is getting deprecated.

I was hoping there would be a way to install specific modules in the pyproject.toml file, but it looks like we can only specify dependency packages. Not sure if we'd then run into import errors given the way we have relative imports set up (through __init__.py) in MHKiT.

[ssolson]

@jmcvey3 was there anything else you wanted to add to #400 besides the upgrade to modern python install?

[jmcvey3]

@ssolson Yeah I thought I had found a way to install particular submodules (e.g., pip install mhkit[wave] only installs packages necessary for the wave module), but I can't seem to find it.

[ssolson]

We could have some common dependencies, then have module specific, and an all modules version like this:

[build-system]
requires = ["setuptools>=61.0", "wheel"]
build-backend = "setuptools.build_meta"

[project]
name = "mhkit"
# `version` is read from `[tools.setuptools.dynamic] version` during build: https://setuptools.pypa.io/en/latest/userguide/pyproject_config.html#dynamic-metadata
dynamic = ["version"]
description = "Marine and Hydrokinetic Toolkit"
readme = "README.md"
authors = [
    {name = "MHKiT developers"}
]
license = {text = "Revised BSD"}
classifiers = [
    "Development Status :: 3 - Alpha",
    "Programming Language :: Python :: 3",
    "Topic :: Scientific/Engineering",
    "Intended Audience :: Science/Research",
    "Operating System :: OS Independent",
]
requires-python = ">=3.10"
dependencies = [
    "numpy>=2.0.0",
    "pandas>=2.2.2",
    "scipy>=1.14.0",
    "xarray>=2024.6.0",
    "matplotlib>=3.9.1",
]

[project.optional-dependencies]
# Core dependencies for each module
wave = [
    "scikit-learn>=1.5.1",
    "statsmodels>=0.14.2",
    "numexpr>=2.10.0",
    "bottleneck",
]

tidal = [
    "scikit-learn>=1.5.1",
    "statsmodels>=0.14.2",
    "numexpr>=2.10.0",
    "bottleneck",
]

river = [
    "scikit-learn>=1.5.1",
    "statsmodels>=0.14.2",
    "numexpr>=2.10.0",
    "bottleneck",
]

dolfyn = [
    "h5py>=3.11.0",
    "h5pyd>=0.18.0",
    "netCDF4>=1.7.1.post1",
    "cartopy",
]

power = [
    "scikit-learn>=1.5.1",
    "statsmodels>=0.14.2",
]

loads = [
    "scikit-learn>=1.5.1",
    "statsmodels>=0.14.2",
]

mooring = [
    "scikit-learn>=1.5.1",
    "statsmodels>=0.14.2",
]

acoustics = [
    "scikit-learn>=1.5.1",
    "statsmodels>=0.14.2",
]

qc = [
    "scikit-learn>=1.5.1",
    "statsmodels>=0.14.2",
]

utils = [
    "requests",
    "beautifulsoup4",
    "lxml",
    "pytz",
    "NREL-rex>=0.2.63",
    "pecos>=0.3.0",
    "fatpack",
]

# Development dependencies
dev = [
    "pytest",
    "pytest-cov",
    "pre-commit",
]

# Install all optional dependencies
all = [
    "mhkit[wave]",
    "mhkit[tidal]",
    "mhkit[river]",
    "mhkit[dolfyn]",
    "mhkit[power]",
    "mhkit[loads]",
    "mhkit[mooring]",
    "mhkit[acoustics]",
    "mhkit[qc]",
    "mhkit[utils]",
]

[project.urls]
Homepage = "https://github.com/MHKiT-Software/mhkit-python"
Documentation = "https://mhkit-software.github.io/MHKiT"

[tool.setuptools]
packages = ["mhkit"]
zip-safe = false
include-package-data = true

[tool.setuptools.dynamic]
version = {attr = "mhkit.__version__"} 

[ssolson]

@jmcvey3 I like this but looking into it more it seems this is not possible with conda (without creating a separate env file for each env.)

We need to determine if we like the added flexibility using pip is worth the inconsistency between the two packaging systems (pip/ conda).

I am slightly leaning towards consistency between the two, but not opposed to allowing the flexibility of module specific installs with pip.

[akeeste]

If we go forward with the flexibility of pip, it's worth adding an example section as well:

# Install all optional dependencies
examples = [
    "folium",
    "utm",
    "jupyter",
    "notebook",
]

I think it's worth trying out the flexible pip approach

[akeeste]

Resolved by #400