RUMM-3475: Fix memory leak in JankStats usage

[0xnm]

What does this PR do?

There is a memory leak in the structure like WeakHashMap<Window, JankStats>, because WeakHashMap has strongly-referenced values, so it means if JankStats holds internally Window instance, it won't be GC-ed, even if it is a key of WeakHashMap.

In this PR we are removing the keys explicitly if activity is destroyed and there are no more activities left. This may increase the number of times we create JankStats instance, but it is not a big deal compared to the memory leak mitigation.

Even if there is a configuration change and activity will be re-created, it may be backed by new/another window anyway.

Review checklist (to be filled by reviewers)

• Feature or bugfix MUST have appropriate tests (unit, integration, e2e)
• Make sure you discussed the feature or bugfix with the maintaining team in an Issue
• Make sure each commit and the PR mention the Issue number (cf the CONTRIBUTING doc)

[codecov-commenter]

Codecov Report

Merging #1553 (c20c477) into develop (01ece69) will decrease coverage by 0.08%.
The diff coverage is 100.00%.

@@             Coverage Diff             @@
##           develop    #1553      +/-   ##
===========================================
- Coverage    83.39%   83.31%   -0.08%     
===========================================
  Files          437      437              
  Lines        14746    14749       +3     
  Branches      2219     2220       +1     
===========================================
- Hits         12297    12287      -10     
- Misses        1868     1872       +4     
- Partials       581      590       +9     

Impacted Files	Coverage Δ
...ernal/vitals/JankStatsActivityLifecycleListener.kt	85.71% <100.00%> (+0.81%)	⬆️

... and 12 files with indirect coverage changes