Fix data saving and display in stat screen

[ymarc11]

This Pull Request refactors how recent speech analysis data is stored and displayed, going from an ArrayDeque to a List for UserStatistics.recentData to make serialization from firestore easier. Alongside this data model change, the problem in the display of the metrics extracted from the data is now fixed and the UI in StatScreen is updated to be scrollable, ensuring that graphs and statistics are more accessible and visually clear.

Data Model Improvements:

• Replaced recentData in UserStatistics from an ArrayDeque to a List and made the conversion 'arrayDeque'.toList() in the UserProfileViewModel.
• Updated UserProfileRepositoryFirestore to correctly extract and map recentData as a list from Firestore (serialization).
• Update getUserProfile to save correctly the data

UI Enhancements:

• Made the StatScreen vertically scrollable (on some screens some composables were hidden).
• Adjusted the values displayed on the y-axis to represent better the intervals of the real values (might change the interval for the talk time seconds y-axis)
• Updated get metric mean to only take two numbers after the decimal for better and cleaner display of the means on the stat screen

Preview:

Related issue:
#273
#297

In this PR , the main bugs of saving and displaying recentData on the statScreen are finally fixed.

[IskandarGhobril]

This PR fixes the stat screen bug and brings solid improvements to both the data model and UI. Great work overall, but I still have a few suggestions for improvement:

• You could introduce a helper function to generate the graphs and their corresponding text dynamically, instead of repeating similar code multiple times.
• The "Talk Time Percentage" graph can be removed since it always displays 100% (based on testing). It simplifies the screen and improves clarity
• We already have a test for documentToUserProfile so adding checks to ensure analysis data is correctly retrieved and mapped from Firestore would not be too hard and it will help reach 80% coverage.
• Consider adding a helper function for converting Firestore data to AnalysisData. This would keep documentToUserProfile cleaner (check out the mapToSpeechBattle function)

Otherwise, great job on addressing the serialization issues, fixing the UI and the display of the stats. Thanks, Marc!

[sonarqubecloud]

Quality Gate passed

Issues
1 New issue
0 Accepted issues

Measures
0 Security Hotspots
72.2% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud

[IskandarGhobril]

Great job on achieving 97% line coverage and incorporating my suggestions! For the graph, it might be best to check with the rest of the team to get their input. I'll go ahead and approve it for now. Thanks again, Marc!

[ymarc11]

Regarding the talkTimePercentage, I will do a new PR for replacing it by a more relevant metric as I as well previously found it to be useless.