consistent use xr.DataArray and xr.Dataset arguments

[jtgrasb]

Description

This pull request makes the exc_coeff to be consistently a DataArray throughout the multiple functions that use it. I also briefly checked some of the other docstrings to confirm inputs and outputs are accurately recorded.

Closes #183

Type of PR

• Bug fix
• New feature
• Documentation
• Other: (specify)

Checklist for PR

• Authors read the contribution guidelines
• The pull request is from an issue branch (not main) on your fork, to the main branch in WecOptTool.
• The authors have given the admins edit access
• All changes adhere to the style guide including PEP8, Docstrings, and Type Hints.
• Modified the documentation if applicable
• Modified or added a new test
• All tests pass
• Reference or close any relevant issues

Additional details

This is in response to issue #183. @akeow I don't know what caused the initial issue so please let me know whether this solves the issue.

[coveralls]

Pull Request Test Coverage Report for Build 6747692411

• 2 of 2 (100.0%) changed or added relevant lines in 1 file are covered.
• No unchanged relevant lines lost coverage.
• Overall coverage remained the same at 91.897%

---

Totals
Change from base Build 6735628848:	0.0%
Covered Lines:	2359
Relevant Lines:	2567

---

💛 - Coveralls

[jtgrasb]

Solve docstring should be fixed now @cmichelenstrofer

[rgcoe]

@jtgrasb - I think this may also need to be updated for the static methods for creating wec objects. For example

WecOptTool/wecopttool/core.py

Lines 511 to 561 in 6c9b852

	@staticmethod
	def from_impedance(
	freqs: ArrayLike,
	impedance: ArrayLike,
	exc_coeff: ArrayLike,
	hydrostatic_stiffness: ndarray,
	f_add: Optional[TIForceDict] = None,
	constraints: Optional[Iterable[Mapping]] = None,
	min_damping: Optional[float] = _default_min_damping,
	) -> TWEC:
	"""Create a WEC object from the intrinsic impedance and
	excitation coefficients.
	The intrinsic (mechanical) impedance :math:`Z(ω)` linearly
	relates excitation forces :math:`F(ω)` to WEC velocity
	:math:`U(ω)` as :math:`ZU=F`.
	Using linear hydrodynamic coefficients, e.g. from a BEM code
	like Capytaine, the impedance is given as
	:math:`Z(ω) = (m+A(ω))*iω + B(ω) + B_f + K/(iω)`.
	The impedance can also be obtained experimentally.
	Note that the impedance is not defined at :math:`ω=0`.
	Parameters
	----------
	freqs
	Frequency vector [:math:`Hz`] not including the zero frequency,
	:python:`freqs = [f1, 2*f1, ..., nfreq*f1]`.
	impedance
	Complex impedance of size :python:`(ndof, ndof, nfreq)`.
	exc_coeff
	Complex excitation transfer function of size
	:python:`(ndof, nfreq)`.
	hydrostatic_stiffness
	Linear hydrostatic restoring coefficient of size
	:python:`(ndof, ndof)`.
	f_add
	Dictionary with entries :python:`{'force_name': fun}`, where
	:python:`fun` has a signature
	:python:`def fun(wec, x_wec, x_opt, waves):`, and returns
	forces in the time-domain of size
	:python:`(2*nfreq, ndof)`.
	constraints
	List of constraints, see documentation for
	:py:func:`scipy.optimize.minimize` for description and
	options of constraints dictionaries.
	If :python:`None`: empty list :python:`[]`.
	min_damping
	Minimum damping level to ensure a stable system.
	See :py:func:`wecopttool.check_impedance` for
	more details.

We should get consistent with this, or (ideally) program in a way that is agnostic to whether the users passes an xr.dataarray or np.ndarray

[jtgrasb]

@ryancoe I changed the title of this PR as it was backwards previously. The PR makes exc_coeff an ArrayLike input wherever it is used. The ArrayLike specification is agnostic to whether the users passes an xr.dataarray or np.ndarray. I think what you're suggesting is to be consistent for other variables such as the hydrostatic_stiffness? If so, I can amend this to make sure that and other variables are all agnostic to array input type.

[rgcoe]

Per discussion today: we like indexing by name, so require the arguments to be xr.DataArrays, etc. After that, merge.