Refactoring 9183, refactor SameString towards more of a case sensitive search

[GaryMarksBigladder]

Pull request overview

• Refactor, for issue Eliminate uses of SameString and transition to case-sensitive string comparisons #9183

Description of the purpose of this PR

Notes on this PR:
To test the code, I ran through all of the test IDFs that I could, a handful of them broke on both develop and my refactor branch. This could be an issue with my testing code. My testing code copied over the IDFs, one at a time to a new folder, it tried to also copy over any additional files, based on the IDF's file name. Once the files were copied over, it ran the IDF 3 times with the built code from the NREL/develop branch and 3 times with my refactor branch (alternating between them, so develop, refactor, develop...). I turned off everything that I could on the computer I was running the tests on, and disconnected it from the network, to try and reduce noise in the timings. After running, the code compared output files to make sure they matched, removing EnergyPlus version numbers and time stamps.

Those tests showed a net improvement of about 1.37% (total develop run time/total refactor run time), and an average improvement of about 0.27% (an average of all average percentage changes). The difference between those is most likely due to minimal performance changes in short running IDFs.

This commit/PR is for the first step in getting that refactor completely done. It should stop the code from doing case insensitive compares more than once. The way it's doing this is by changing the casing when SameString finds equality but with different cases.

The second step in this would be to only uppercase (vs matching the case of what it compared to), and log every time it changes the case to try to make certain the code isn't using multiple cases with-in itself.

This commit also tightens up the enum check by making that case insensitive, using equali, so all the makeUPPER wraps being done before that can be safely removed. I did that because there were inconsistencies in some cases/if elses, where most of the time makeUPPER was being used before the enum check, but several times it was forgotten. This should fix that, and hopefully fix some lurking bugs along the way.

I also sped up SameString/equali so there's less of a hit with the case insensitive searching, often making it faster than a case sensitive equality check. I did this by adding a middle character check to exit the comparison quickly if the middle character was different. This came from looking at the data we're checking and seeing that there were many cases where a large chunk of the beginning of the string, and a small chunk of the end of the string were often the same with the middle being different. This could probably be optimized more, but I'm not certain we would get much performance improvement. This change, and adding this to the enum check is where I saw the greatest performance improvement.

Pull Request Author

• Title of PR should be user-synopsis style (clearly understandable in a standalone changelog context)
• Label the PR with at least one of: Defect, Refactoring, NewFeature, Performance, and/or DoNoPublish
• Pull requests that impact EnergyPlus code must also include unit tests to cover enhancement or defect repair (NA, I did run the refactor against EnergyPlus's test IDFs)
• Author should provide a "walkthrough" of relevant code changes using a GitHub code review comment process

Reviewer

• Perform a Code Review on GitHub
• If branch is behind develop, merge develop and build locally to check for side effects of the merge
• If defect, verify by running develop branch and reproducing defect, then running PR and reproducing fix
• If feature, test running new feature, try creative ways to break it
• CI status: all green or justified
• Check that performance is not impacted (CI Linux results include performance check)
• Run Unit Test(s) locally
• Check any new function arguments for performance impacts
• Verify IDF naming conventions and styles, memos and notes and defaults
• If new idf included, locally check the err file and other outputs

Notes on this PR:
To test the code, I ran through all of the test IDFs that I could, a handful of them broke on both develop and my refactor branch. This could be an issue with my testing code. My testing code copied over the IDFs, one at a time to a new folder, it tried to also copy over any additional files, based on the IDF's file name. Once the files were copied over, it ran the IDF 3 times with the built code from the NREL/develop branch and 3 times with my refactor branch (alternating between them, so develop, refactor, develop...). I turned off everything that I could on the computer I was running the tests on, and disconnected it from the network, to try and reduce noise in the timings. After running, the code compared output files to make sure they matched, removing EnergyPlus version numbers and time stamps.

Those tests showed a net improvement of about 1.37% (total develop run time/total refactor run time), and an average improvement of about 0.27% (an average of all average percentage changes). The difference between those is most likely due to minimal performance changes in short running IDFs.

This commit/PR is for the first step in getting that refactor completely done. It should stop the code from doing case insensitive compares more than once. The way it's doing this is by changing the casing when SameString finds equality but with different cases.

The second step in this would be to only uppercase (vs matching the case of what it compared to), and log every time it changes the case to try to make certain the code isn't using multiple cases with-in itself.

This commit also tightens up the enum check by making that case insensitive, using equali, so all the makeUPPER wraps being done before that can be safely removed. I did that because there were inconsistencies in some cases/if elses, where most of the time makeUPPER was being used before the enum check, but several times it was forgotten. This should fix that, and hopefully fix some lurking bugs along the way.

I also sped up SameString/equali so there's less of a hit with the case insensitive searching, often making it faster than a case sensitive equality check. I did this by adding a middle character check to exit the comparison quickly if the middle character was different. This came from looking at the data we're checking and seeing that there were many cases where a large chunk of the beginning of the string, and a small chunk of the end of the string were often the same with the middle being different. This could probably be optimized more, but I'm not certain we would get much performance improvement. This change, and adding this to the enum check is where I saw the greatest performance improvement.

[mitchute]

I think we'll want to use .empty() for these.