[Bug] Incorrectly including tables matching excludingTablePattern in combined mode cdc

### Search before asking

- [X] I searched in the [issues](https://github.com/apache/paimon/issues) and found nothing similar.

### Paimon version

master

### Compute Engine

flink

### Minimal reproduce step

1. Create a database named cdc_test in mysql, then create primary table named pk_1, pk_2 ... pk_100 and non-primary table named non_pk_1, non_pk_2, ... non_pk_100.
2. Start a combined mode mysql database cdc job, and set excludingTablePattern to 'non_pk_.+'.
3. Concurently create non-primary table named non_pk_101 when starting cdc job.
4. Finally, the Jobmanager log will show "com.ververica.cdc.connectors.mysql.source.utils.TableDiscoveryUtils [] - including ‘cdc_test.non_pk_101’ for further processing".

### What doesn't meet your expectations?

non_pk_101 obviously matches the excludingTablePattern and needs to be excluded. Since non_pk_101 is a non-primary key table, MySqlChunkSplitter will report an error:
Caused by: org.apache.flink.table.api.ValidationException: Incremental snapshot for tables requires primary key, but table cdc_test.non_pk_101 doesn't have primary key.

combinedModeTableList func uses excluding pattern ?!(^db\\.tbl$)|(^...$)，and this will miss the newly table that matches excludingTablePattern and created when starting cdc job.

### Anything else?

_No response_

### Are you willing to submit a PR?

- [X] I'm willing to submit a PR!

Start by tracing the combinedModeTableList function and the TableDiscoveryUtils log path involved in MySQL CDC table discovery. Reproduce the concurrent creation of non_pk_101 with excludingTablePattern set, then verify that the matching table is excluded and MySqlChunkSplitter no longer reports the missing-primary-key error.

Written by the indexing model from the issue text.