feat: deposit_v4 with cheaper depositTopUp

[86667]

Work done by updateLatestComputedEpoch() within functions called by delegators, ie depositTopUp(), unstake() and withdraw() is expensive in some circumstances, ie when a deposit has occurred and one or more epochs passed since. This is because the new staker key is written to future committees.

This work aims to remove costly writes to _committee[] from the above functions and instead have all writes done in deposit(). To achieve this i've changed how _committee works slightly such that we only write to _committee[] if there are changes and do not copy data if data hasnt changed.

Results

Before

deposit owner1: Gas used 410503

Rolling ahead 2 epochs
deposit owner2: Gas used 644369

Rolling ahead 1 epochs
1st top up owner 1: Gas used 179666
1st top up owner 2: Gas used 16149

Rolling ahead 2 epochs
2nd top up owner 1: Gas used 273779
2nd top up owner 2: Gas used 16158

After

deposit owner1: Gas used 334749

Rolling ahead 2 epochs
deposit owner2: Gas used 446585

Rolling ahead 1 epochs
1st top up owner 1: Gas used 14174
1st top up owner 2: Gas used 14174

Rolling ahead 2 epochs
2nd top up owner 1: Gas used 13987
2nd top up owner 2: Gas used 13990

[github-actions]

Bencher Report

Branch	1991-incorrect-estimation-and-high-gas-costs-on-first-call-to-deposittopupdeposit
Testbed	self-hosted

⚠️ WARNING: No Threshold found!

Without a Threshold, no Alerts will ever be generated.

• Latency (nanoseconds (ns))

Click here to create a new Threshold
For more information, see the Threshold documentation.
To only post results if a Threshold exists, set the --ci-only-thresholds CLI flag.

Click to view all benchmark results

Benchmark	Latency	milliseconds (ms)
process-empty/process-empty	📈 view plot ⚠️ NO THRESHOLD	9.63
produce-full/produce-full	📈 view plot ⚠️ NO THRESHOLD	1,964.70

🐰 View full continuous benchmarking report in Bencher

[86667]

Migrated along with another solution to a single draft PR - #2226