DF's behaviour is unexpected when using of excel's stream reading

Hi!

Current Behavior

I am trying to laod .xlsx file into dataframe. The size of the file is about 40Mb. Putting data to DF goes fine but when I invoke any method on df it give me no result for a long time(more than an hour) and I cancel a job.

Steps to Reproduce (for bugs)

the first step looks like this and works succesfully
```
val sparkDF = spark.read.format("com.crealytics.spark.excel")
            .option("header", true)
            .option("inferSchema", "true")
             .option("excerptSize", 100)
           .option("maxRowsInMemory", 2)
            .option("sheetName", "OSA")
            .load("/mnt/databricks/TEST/KPI_list_082020.xlsx")
```

The next step is counting rows: sparkDF.count(),show(). Regardless of maxRowsInMemory parameter - 1 and up to 20, it works the same - I have no results for hours.

Your Environment

I am using Databricks cluster with the following configuration ( Apache Spark 2.4.5, Scala 2.11) Worker Type 14.0 GB Memory, 4 Cores, 0.75 DBU Driver Type 14.0 GB Memory, 4 Cores, 0.75 DBU

File is placed in Azure Blob Storage.

[vasilnikolay]

@EnverOsmanov , It works for 20 seconds for 40Mb(~600k rows). Thanks a lot!

[EnverOsmanov]

@vasilnikolay , right. Yes, you can. But it is much easier to use 0.13.5. It is available since my last comment.