Bug #58896 MTR should recognise combinations as experimental without needing wildcards

Added a pattern match to cover combinations
Added to readme file

mysql-test/collections/README.experimental

@@ -15,9 +15,13 @@ The syntax is as follows:
    and any subsequent characters are ignored.
 
 4) The full test case name including the suite and execution mode
-   must be specified, for example:
+   may be specified, for example:
         main.alias 'row'     # bug#00000
 
+4b) Now, combinations will also be covered if only the test name is
+    specified, for example:
+        rpl.rpl_ps       # Covers 'row', 'mix' and 'stmt'
+
 5) As an exception to item 4, the last character of the test case
    specification may be an asterisk (*). In that case, all test cases that
    start with the same characters up to the last letter before the asterisk

mysql-test/collections/default.experimental

@@ -21,7 +21,7 @@ main.outfile_loaddata @solaris           # joro : Bug #46895
 
 ndb.*                                    # joro : NDB tests marked as experimental as agreed with bochklin
 
-rpl.rpl_innodb_bug28430*  @solaris       # Bug#46029
+rpl.rpl_innodb_bug28430   @solaris       # Bug#46029
 rpl.rpl_row_sp011         @solaris       # Joro : Bug #54138
 
 rpl_ndb.*                                # joro : NDB tests marked as experimental as agreed with bochklin

mysql-test/lib/mtr_report.pm

@@ -129,7 +129,8 @@ sub mtr_report_test ($) {
       # Find out if this test case is an experimental one, so we can treat
       # the failure as an expected failure instead of a regression.
       for my $exp ( @$::experimental_test_cases ) {
-        if ( $exp ne $test_name ) {
+	# Include pattern match for combinations
+        if ( $exp ne $test_name && $test_name !~ /^$exp / ) {
           # if the expression is not the name of this test case, but has
           # an asterisk at the end, determine if the characters up to
           # but excluding the asterisk are the same