#240 Clarify normality test label description

[Artur Tarassow]

The following message was posted by user "blbcava" on April 8 on github:
https://github.com/gretl-project/gretl/issues/3

In informed the user that I re-post it here:

<>
When I run OLS and look at the residual normality test in the model output, it just says:

Test for normality of residual:
Null hypothesis: error is normally distributed
Test statistic: Chi-square(2) = 6.533 [0.0381]

No mention that this is Doornik-Hansen. I spent some time confused because Python's statsmodels gives 6.112 for the same residuals - turns out that's the classical Jarque-Bera, not DH. Same null, same chi²(2), but different statistics.

Looking at the source, the standalone normality command already prints "Doornik-Hansen test" correctly (describe.c ~line 1650). The issue is in the tstrings[] array in gretl_model.c (~line 4183):

{ GRETL_TEST_NORMAL,
N_("Test for normality of residual"),
N_("error is normally distributed") },

Changing the first string to "Doornik-Hansen test for normality of residual" would be enough - no computed values change, just the label. Would make the model output consistent with what normality already prints.
</>