This is for issue pingcap#1147

When creating a changefeed with `--opts nameStrategy=topic` this now
uses the topic name as basis for the schema registry. Before this PR
the schema registry always used `{schema}_{table}` as basis.

This solves the case where KSQL expect the topic name to be used as
basis for the schema registry.

== Example

Setup a test environment and create the changefeed

```
tiup playground --without-monitor --tiflash 0 v5.3.0
confluent local services start
./bin/cdc server
./bin/cdc cli changefeed create --no-confirm \
--changefeed-id="simple-replication-task" \
--sort-engine="unified" \
--sink-uri="kafka://127.0.0.1:9092/cdctest?protocol=avro&kafka-version=2.8.0" \
--opts registry="http://127.0.0.1:8081" \
--opts nameStrategy=topic
```

To verify the `nameStrategy`:
```
$ ./bin/cdc cli changefeed query -c simple-replication-task | jq '.info.opts.nameStrategy'
"topic"
```

Now let's create a table that uses this changefeed:
```
CREATE TABLE t1 (id INT PRIMARY KEY, n VARCHAR(200) NOT NULL, ts TIMESTAMP(6) NOT NULL DEFAULT CURRENT_TIMESTAMP(6));
INSERT INTO t1(id, n) VALUES (1, 'test');
```

This now shows up in KSQL like this:
```
ksql> SHOW TOPICS;

 Kafka Topic                 | Partitions | Partition Replicas
---------------------------------------------------------------
 cdctest                     | 3          | 1
 default_ksql_processing_log | 1          | 1
---------------------------------------------------------------
ksql> PRINT cdctest FROM BEGINNING;
Key format: AVRO or KAFKA_STRING
Value format: AVRO
rowtime: 2021/12/17 07:06:07.227 Z, key: {"id": 1}, value: {"id": 1, "n": "test", "ts": 1639724765242}, partition: 0
^CTopic printing ceased
```

And in the schema registry:
```
$ curl -s http://127.0.0.1:8081/subjects | jq
[
  "cdctest-key",
  "cdctest-value"
]
```

This now allows us to do this:
```
ksql> CREATE STREAM t1str WITH (KAFKA_TOPIC='cdctest', VALUE_FORMAT='avro');

 Message
----------------
 Stream created
----------------

ksql> DESCRIBE t1str;

Name                 : T1STR
 Field | Type
-------------------------
 ID    | INTEGER
 N     | VARCHAR(STRING)
 TS    | TIMESTAMP
-------------------------
For runtime statistics and query details run: DESCRIBE <Stream,Table> EXTENDED;
```

Note that we didnt' have to specify the columns in the `CREATE STREAM`
comman as these were fetched from the schema registry.

See also:
- https://docs.ksqldb.io/en/latest/operate-and-deploy/schema-registry-integration/#schema-inference