ParquetAvro int96 parity with parquetNative reader

[swaminathanmanish]

Whats in the PR:
Keeping int96 interpretation the same for parquetAvro and parquetNative readers.

Why its needed:
This PR allowed int96 to be read by parquetAvro reader. However it is read as byte array (12 bytes). The parquetNative extractor translates int96 to a long. This PR does the same translation when parquetAvro reader is used.

Test:
parquetNative and ParquetAvro output values should match for int96 column.

[codecov-commenter]

Codecov Report

Attention: Patch coverage is 92.00000% with 2 lines in your changes are missing coverage. Please review.

Project coverage is 61.78%. Comparing base (59551e4) to head (15e17e1).
Report is 11 commits behind head on master.

Files	Patch %	Lines
...nputformat/parquet/ParquetAvroRecordExtractor.java	87.50%	1 Missing and 1 partial ⚠️

Additional details and impacted files

@@             Coverage Diff              @@
##             master   #12496      +/-   ##
============================================
+ Coverage     61.75%   61.78%   +0.03%     
  Complexity      207      207              
============================================
  Files          2436     2437       +1     
  Lines        133233   133255      +22     
  Branches      20636    20644       +8     
============================================
+ Hits          82274    82328      +54     
+ Misses        44911    44874      -37     
- Partials       6048     6053       +5     

Flag	Coverage Δ
custom-integration1	<0.01% <0.00%> (ø)
integration	<0.01% <0.00%> (ø)
integration1	<0.01% <0.00%> (ø)
integration2	0.00% <0.00%> (ø)
java-11	61.70% <92.00%> (-0.01%)	⬇️
java-21	61.66% <92.00%> (+0.04%)	⬆️
skip-bytebuffers-false	61.75% <92.00%> (+0.01%)	⬆️
skip-bytebuffers-true	61.60% <92.00%> (+33.88%)	⬆️
temurin	61.78% <92.00%> (+0.03%)	⬆️
unittests	61.77% <92.00%> (+0.03%)	⬆️
unittests1	46.90% <0.00%> (+0.01%)	⬆️
unittests2	27.75% <92.00%> (+0.02%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[snleee]

LGTM