optimize cpu_profile_transformer to not create intermediate lists

[jakemac53]

Uses an iterator over the stack frame values in CpuProfileTransformer, instead of allocating new lists for they keys and values in the map, which are then read by index. I also removed the late/nullable fields in favor of local variables, since the state was never used outside of a single method call.

As far as I could tell the keys were never actually used except to look up values in the map, so I dropped that entirely. There was an extra lookup in the map and local variable which I am pretty certain wasn't needed.

This isn't directly related to any specific issue, I just noticed it as I was perusing through some of the code for the cpu profiler and figured I would send something out, to dip my toes in here :).

No tests were added, the behavior should be unchanged so hopefully existing tests should cover it.

I did also add a TODO because I noticed a piece of logic that was O(N^2) worst case here but I don't know enough about how large the N is expected to be or whether its a common path, and it would be more involved to fix.

Pre-launch Checklist

• I read the [Contributor Guide] and followed the process outlined there for submitting PRs.
• I read the [Tree Hygiene] wiki page, which explains my responsibilities.
• I read the [Flutter Style Guide] recently, and have followed its advice.
• I signed the [CLA].
• I listed at least one issue that this PR fixes in the description above.
• I updated/added relevant documentation (doc comments with ///).
• I added new tests to check the change I am making, or there is a reason for not adding tests.

![build.yaml badge]

[kenzieschmoll]

Related to #7917, which enumerates several performance improvements we should make to the CPU profiler code.