Update benchmark results for version v3.0.0

[meet2mky]

Description

Update benchmarking numbers for GCSFuse Version v3.0.0 at tag v3.0.0

• For each read benchmark metadata cache was prefetched using ls -R "$MOUNT_POINT"
• For each write benchmark we ran it twice to ensure first benchmark creates files and then we delete files to ensure negative stat cache and then results of second write benchmarks were taken.
• Converted GiB and MiB to GB and MB for bandwidth calculation.
• Used numjob=48 for 1G write benchmarks instead of numjob=112 (Ref: Update public benchmarks for gcsfuse version v3.0.0 #3399)

Link to the issue in case of a bug fix.

b/415727864

Testing details

1. Manual - NA
2. Unit tests - NA
3. Integration tests - NA

Any backward incompatible change? If so, please explain.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 78.85%. Comparing base (9a7bdf7) to head (a2651f7).
Report is 8 commits behind head on master.

Additional details and impacted files

@@           Coverage Diff           @@
##           master    #3408   +/-   ##
=======================================
  Coverage   78.85%   78.85%           
=======================================
  Files         137      138    +1     
  Lines       18313    18324   +11     
=======================================
+ Hits        14440    14450   +10     
- Misses       3389     3390    +1     
  Partials      484      484           

Flag	Coverage Δ
unittests	78.85% <ø> (+<0.01%)	⬆️

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.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[charith87]

Please add if any new steps are added below in steps to benchmark. I see its mentioned you had prefetch etc. good to mention in the steps so customers can redo it as well.

[meet2mky]

These benchmarks are not reproducible at a single row level so even if mention it there may be variations, We have logged a bug (b/423764522) to fix this in future. Let me know if we need to do for V3.0.0 benchmark update as it was decided in a thread to not do for V3.0.0
CC(folks who are involved in the discussion): @mustvicky @anushka567 @ashmeenkaur

[meet2mky]

I think, we can add the conditions for the benchmarks though. wdyt? @anushka567

[charith87]

Also assuming this bump up is due to GiB to GB conversion for other ones. Here it is due to streaming writes.
Hoping the numbers are validated by @ashmeenkaur or @vadlakondaswetha