Potential performance issue: Unreliable performance of pd.read_excel in pandas 1.2.4

[TendouArisu]

Issue Description:

Hello.
I have discovered a performance degradation in the pd.read_excel function of pandas version 1.2.4 when pd.read_excel read_excel trailing blank rows and columns. And I notice that requirements.txt, shows that it depends on pandas version 1.2.4. I am not sure whether this performance problem in pandas will affect this repository. I found some discussions on GitHub related to this issue, including #41167 and #41227.
I also found that tools/excel_tools.py used the influenced api. There may be more files used the influenced api.

Suggestion

I would recommend considering an upgrade to a different version of pandas > 1.2.4 or exploring other solutions to optimize the performance.
Any other workarounds or solutions would be greatly appreciated.
Thank you!