[SPARK-56021][SS] Increase AutoSnapshotRepair default maxChangeFileReplay threshold from 50 to 500

### What changes were proposed in this pull request?

Increasing the default from 50 to 500.

### Why are the changes needed?

50 is too conservative. Users would prefer to spend more time repairing the pipeline than having the pipeline completely down.

### Does this PR introduce _any_ user-facing change?

No

### How was this patch tested?

existing tests

### Was this patch authored or co-authored using generative AI tooling?

Claude-4.5-opus

Closes #54843 from micheal-o/auto_repair_max_changefile_increase.

Authored-by: micheal-o <micheal.okutubo@gmail.com>
Signed-off-by: Anish Shrigondekar <anish.shrigondekar@databricks.com>

Author: micheal-o

sql/catalyst/src/main/scala/org/apache/spark/sql/internal/SQLConf.scala

@@ -2717,7 +2717,7 @@ object SQLConf {
       .version("4.1.0")
       .intConf
       .checkValue(k => k > 0, "Must allow at least 1 change file replay")
-      .createWithDefault(50)
+      .createWithDefault(500)
 
   val STATE_STORE_INSTANCE_METRICS_REPORT_LIMIT =
     buildConf("spark.sql.streaming.stateStore.numStateStoreInstanceMetricsToReport")