Shorten the list of supported argument types in docstrings

[seisman]

I'll take the spacing parameter as an example, which is used in the xyz2grd and many other functions.

Here is the source code of the spacing docstrings:

pygmt/pygmt/helpers/decorators.py

Lines 300 to 302 in 5b4d5c3

	"spacing": r"""
	spacing : int or float or str or list or tuple
	*x_inc*\ [**+e**\|\ **n**][/\ *y_inc*\ [**+e**\|\ **n**]].

Here is what it looks like:

As you can see, currently, spacing is documented to support 5 types: int, float, str, list and tuple.

My thoughts:

1. Simplify int or float to float. If an argument can accept float (e.g., 1.0), it definitely can accept int (e.g., 1). It's not necessary to list both.
2. Simplify list or tuple to list or tuple or something else.

[yvonnefroehlich]

1. Simplify int or float to float. If an argument can accept float (e.g., 1.0), it definitely can accept int (e.g., 1). It's not necessary to list both.

I aggree, if float is accepted, it is clear also int can be passed.

2. Simplify list or tuple to list or tuple or something else.

I believe we already often only mention list in the documentation, even though "tuple" is also allowed, and use list more often in the examples (?). Thus, if we summarize list or tuple to one term, I think we should use list.

A third point, which is a bit related to this: We write inconsistently or (e.g., pygmt.xyz2grd) or commas (e.g., pygmt.Figure.meca) between more than two input types. I feel it would be nice to have this conistent. The comma version seem to be a bit shorter.

[seisman]

A third point, which is a bit related to this: We write inconsistently or (e.g., pygmt.xyz2grd) or commas (e.g., pygmt.Figure.meca) between more than two input types. I feel it would be nice to have this conistent. The comma version seem to be a bit shorter.

That's a good point. I also prefer the comma version, which is also used by pandas (e.g., https://pandas.pydata.org/docs/reference/api/pandas.read_csv.html)