Improve the traversing of snapshot files

[aivinog1]

Is your feature request related to a problem? Please describe.

I think that it is not a good idea to collect snapshot files in the list.
Check it here:

zeebe/snapshot/src/main/java/io/camunda/zeebe/snapshots/impl/SnapshotChecksum.java

Line 41 in 02d1df8

final SfvChecksum sfvChecksum = createCombinedChecksum(fileStream.toList());

.
Because, after a while, the amount of snapshot files should grow. So, we could instead call the .forEachOrdered method to calculate the snapshot. I will provide the PR soon to see this in action and benchmark this.

Describe the solution you'd like
We should call Stream.forEachOrdered instead of collecting snapshots Files in the list.

Describe alternatives you've considered
We could use Stream.forEach but as I can understand the order is important, so we shouldn't do this.

[aivinog1]

@Zelldon Hello 👋
I think that this could be addressed to the ZDP Team (sorry if I am wrong 😅)
Could we assign this to someone, please? In our load test environment, we see a little degradation in creating snapshots and we are suspecting this part of the code.
Thank you in advance 🙂

[Zelldon]

Looks like @deepthidevaki will look into your PR, thanks for providing it :)