CLR did not use na.rm

[TuomasBorman]

No description provided.

[TuomasBorman]

Fixing issue #661

[jarioksa]

Looks OK and safe, but do not change the DESCRIPTION: we change version numbers only at CRAN releases.

[jarioksa]

At closer look, there seems to be a deeper problem in "clr", "alr" and "rclr". None of these passes decostand(..., na.rm) argument to the corresponding .calc_* function. Therefore user cannot set na.rm action, but the value na.rm = TRUE of all .calc_* functions is de facto hard-coded. What you found out was that this setting was not used in one case, but changing this won't allow users to set na.rm in the decostand call (or vegdist call).

I won't merge this PR because much larger changes are needed to pass decostand and vegdist arguments.

[jarioksa]

I think that the problem is bigger than this one, but your fix will be a part of the solution even if larger changes are needed. Thanks for your contribution!