test: add baseline test (#3070)

test/RunTests

@@ -2437,4 +2437,8 @@ train-sets/ref/slates_simple_w_interactions.stderr
     train-sets/ref/cb_dro_explore_adf_sm.stderr
     pred-sets/ref/cb_dro_explore_adf_sm.predict
 
+# Test 335: test --baseline
+{VW} -d train-sets/0001.dat --baseline
+    train-sets/ref/baseline_test.stderr
+
 # Do not delete this line or the empty line above it

test/train-sets/ref/baseline_test.stderr

@@ -0,0 +1,27 @@
+Num weight bits = 18
+learning rate = 0.5
+initial_t = 0
+power_t = 0.5
+using no cache
+Reading datafile = train-sets/0001.dat
+num sources = 1
+Enabled reductions: gd, baseline, scorer
+average  since         example        example  current  current  current
+loss     last          counter         weight    label  predict features
+1.000000 1.000000            1            1.0   1.0000   0.0000       51
+0.648807 0.297614            2            2.0   0.0000   0.5455      104
+0.397826 0.146845            4            4.0   0.0000   0.3158      135
+0.295379 0.192932            8            8.0   0.0000   0.2760      146
+0.253704 0.212029           16           16.0   1.0000   0.4438       24
+0.237305 0.220907           32           32.0   0.0000   0.3310       32
+0.231645 0.225984           64           64.0   0.0000   0.1647       61
+0.221327 0.211009          128          128.0   1.0000   0.8441      106
+
+finished run
+number of examples = 200
+weighted example sum = 200.000000
+weighted label sum = 91.000000
+average loss = 0.190245
+best constant = 0.455000
+best constant's loss = 0.247975
+total feature number = 15482