Fix handling of zero weights

[nalimilan]

Observations with zero weights have to be skipped for probability and analytic weights. Values have been checked against R. Avoid duplication by using a single method for all LinPredModels (though this means we cannot call nobs from functions which don't get passed the full model).

Also fix name of internal function link, which actually returns the distribution.

Improves on #487. Cc: @gragusa

[nalimilan]

@gragusa I get slightly different standard errors with svyglm. Is this expected? Generally in my tests I got exactly the same values up to a least four decimals.

> X = 1:10
> y = c(1, 4, 6, 2, 3, 5, 6, 7, 1, 6)
> wts = c(1, 0, 4, 0, 5, 2, 6, 4, 2, 6)
> svyd <- svydesign(~1, weights=wts, data=data.frame(X=X, y=y, wts=wts))

> summary(svyglm(y ~ X, svyd))

Call:
svyglm(formula = y ~ X, design = svyd)

Survey design:
svydesign(~1, weights = wts, data = data.frame(X = X, y = y, 
    wts = wts))

Coefficients:
            Estimate Std. Error t value Pr(>|t|)  
(Intercept)   3.6707     1.7371   2.113    0.079 .
X             0.2073     0.2313   0.896    0.405  
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

(Dispersion parameter for gaussian family taken to be 3.500116)

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 96.96%. Comparing base (17a568c) to head (add9660).

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #638      +/-   ##
==========================================
+ Coverage   96.94%   96.96%   +0.01%     
==========================================
  Files           8        8              
  Lines        1213     1219       +6     
==========================================
+ Hits         1176     1182       +6     
  Misses         37       37              

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[devmotion]

Isn't this change unnecessary since nobs is fixed below?

Suggested change

	return wts isa ProbabilityWeights ? d * sum(!iszero, wts) / sum(wts) : d
	return wts isa ProbabilityWeights ? d * nobs(r) / sum(wts) : d

[nalimilan]

That was because I defined the method only on LinPredModel and not on the response object. But I've discovered that we have ModResp that covers both LMs and GLMs so I've changed the method.