[Bug] Csv import meet error "There are too many timeseries in memory"

### Search before asking

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

### Version

1.0.0

### Describe the bug and provide the minimal reproduce step

Meet this error when import data to iotdb(version: 1.0.0):
"Meet error when insert csv because 507: There are too many timeseries in memory, please increase MAX_HEAP_SIZE in datanode-env.sh/bat, restart and create timeseries again."

Are there any other options or configurations to avoid this as for now i can`t increase memory size of the server? I don't understand why there are so many timeseries in memory, as there are no queries at the moment. If these timeseries in memory are cache of imported data, when will the memory will be released?

### What did you expect to see?

-

### What did you see instead?

-

### Anything else?

_No response_

### Are you willing to submit a PR?

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

Start by reproducing the CSV import failure on IoTDB 1.0.0 and inspect the timeseries-memory limit referenced in datanode-env.sh/bat. Determine why the import reaches the limit without queries and whether memory is released; done means documenting the cause and a supported configuration or workaround for servers that cannot increase heap size.

Written by the indexing model from the issue text.