meca: inaccurate and confusing parameter names for "PRINCIPAL_AXIS" convention

[seisman]

Description of the problem

The "PRICIPAL_AXIS" convention has 10 parameters: the value (in 10*exponent dynes-cm), azimuth, and plunge of N, T, and P axis, and also the exponent of the values. See https://docs.generic-mapping-tools.org/latest/supplements/seis/meca.html#s for the official documentation. Currently, the meca function defines the following parameters:

pygmt/pygmt/src/meca.py

Lines 255 to 266 in 4b80f3c

	"PRINCIPAL_AXIS": [
	"t_exponent",
	"t_azimuth",
	"t_plunge",
	"n_exponent",
	"n_azimuth",
	"n_plunge",
	"p_exponent",
	"p_azimuth",
	"p_plunge",
	"exponent",
	],

The parameter name t_exponent, n_exponent, and p_exponent is inaccurate and confusing, and should be renamed to t_value, n_value, and p_value.

I think we should make the change in a backward-compatible way and it should go through the deprecation cycle.

[seisman]

I'm closing the issue without writing codes to make it backward-compatible. The reasons are:

1. The meca function was rewritten in PR Rewrite the meca function to support offsetting and labeling beachballs #1784, which already brought some incompatible changes
2. The PRICICIPAL_AXIS convention is not commonly used.
3. It's a bug and was fixed in Rewrite the meca function to support offsetting and labeling beachballs #1784, and we don't have to keep backward-compatibility with buggy behavior.

Feel free to reopen it if anyone insists that backward-compatibility is necessary for it.