[SPARK-9492] [ML] [R] LogisticRegression in R should provide model statistics

[yanboliang]

Like ml LinearRegression, LogisticRegression should provide a training summary including feature names and their coefficients.

[yanboliang]

This PR depends on #9302 , after it merged I will add this test case which has passed on my local test environment.

test_that("summary coefficients match with native glm of Logistic Regression", {
  df <- createDataFrame(sqlContext, iris)
  training <- filter(df, df$Species != "setosa")
  stats <- summary(glm(Species ~ Sepal_Length + Sepal_Width, data = training, family = "binomial"))
  coefs <- as.vector(stats$coefficients)

  rTraining <- iris[iris$Species %in% c("versicolor","virginica"),]
  rCoefs <- as.vector(coef(glm(Species ~ Sepal.Length + Sepal.Width, data = rTraining,
    family = binomial(link = "logit"))))

  expect_true(all(abs(rCoefs - coefs) < 1e-4))
  expect_true(all(
    as.character(stats$features) ==
    c("(Intercept)", "Sepal_Length", "Sepal_Width")))
})

[SparkQA]

Test build #44438 has finished for PR 9303 at commit 9261dcb.

• This patch fails MiMa tests.
• This patch merges cleanly.
• This patch adds no public classes.

[SparkQA]

Test build #44511 has finished for PR 9303 at commit 56bd22d.

• This patch passes all tests.
• This patch merges cleanly.
• This patch adds no public classes.

[mengxr]

@yanboliang Could you rebase master?

[SparkQA]

Test build #44995 has finished for PR 9303 at commit af3e7c7.

• This patch fails Spark unit tests.
• This patch merges cleanly.
• This patch adds no public classes.

[yanboliang]

Jenkins, test this please.

[SparkQA]

Test build #45005 has finished for PR 9303 at commit af3e7c7.

• This patch passes all tests.
• This patch merges cleanly.
• This patch adds the following public classes (experimental):\n * case class ExecutorLostFailure(\n

[mengxr]

LGTM. Merged into master. Thanks!