inconsistencies and typos in bin/diagnose utility output

[jgabry]

1. typo in split r-hat message:

Split R-hat values satisfactory all parameters.

It's missing "for". It should be

Split R-hat values satisfactory for all parameters.

2. inconsistent language for ess and r-hat:

Effective sample size satisfactory.
Split R-hat values satisfactory all parameters.

Both r-hat and ess are for "all parameters" so either they should both say "for all parameters" or neither should

3. inconsistent language for treedepth and divergences:

Checking sampler transitions treedepth.
Treedepth satisfactory for all transitions.

Checking sampler transitions for divergences.
No divergent transitions found.

These both refer to counting how many times something happened (a divergence or hitting max treedepth) so if we're going to say "No divergent transitions found" we should say "No saturation of max treedepth found" or something like that (not sure what the best terminology is) or we should use the "satisfactory" language for both. Right now it's just inconsistent.