Column statistics are not collected when the column name is escaped

### Apache Iceberg version

1.7.1 (latest release)

### Query engine

None

### Please describe the bug 🐞

When an Iceberg has many columns, it calculates statistics only up to 100 columns (can be changed using config `write.metadata.metrics.max-inferred-column-defaults`).

When a column name is escaped, there is a bug that it tries to use the escaped column name when calculating the statistics while the metrics mode was saved using the original column name.

For example, I have a column `$event_time` which is escaped to `_x24event_time`.
When building the metrics mode, it uses the original name in [MetricsConfig.java:138](https://github.com/apache/iceberg/blob/main/core/src/main/java/org/apache/iceberg/MetricsConfig.java#L138):

while when closing the parquet writer, it tries to get the metrics mode for `_x24event_time` in [MetricsUtil.java:118](https://github.com/apache/iceberg/blob/main/core/src/main/java/org/apache/iceberg/MetricsUtil.java#L118):


when it doesn't find the metrics mode, it uses the default which is `none` so the stats are not collected.

### Willingness to contribute

- [ ] I can contribute a fix for this bug independently
- [ ] I would be willing to contribute a fix for this bug with guidance from the Iceberg community
- [x] I cannot contribute a fix for this bug at this time