Include DROPPED_COLUMNS in schema digest computation.

CASSANDRA-11050 introduced dropped columns but excluded it from schema digest computations, with
a comment to include the table in the next major release - 4.0.

The patch now includes DROPPED_COLUMNS in the schema digest computation.

Patch by Jon Meredith; reviewed by Aleksey Yeschenko for CASSANDRA-15843

CHANGES.txt

 4.0-alpha5
+ * Include DROPPED_COLUMNS in schema digest computation (CASSANDRA-15843)
  * Fix Cassandra restart from rpm install (CASSANDRA-15830)
  * Improve handling of 2i initialization failures (CASSANDRA-13606)
  * Add completion_ratio column to sstable_tasks virtual table (CASANDRA-15759)

src/java/org/apache/cassandra/schema/SchemaKeyspace.java

@@ -355,11 +355,6 @@ public final class SchemaKeyspace
         Digest digest = Digest.forSchema();
         for (String table : ALL)
         {
-            // Due to CASSANDRA-11050 we want to exclude DROPPED_COLUMNS for schema digest computation. We can and
-            // should remove that in the next major release (so C* 4.0).
-            if (table.equals(DROPPED_COLUMNS))
-                continue;
-
             ReadCommand cmd = getReadCommandForTableSchema(table);
             try (ReadExecutionController executionController = cmd.executionController();
                  PartitionIterator schema = cmd.executeInternal(executionController))