Resolves #8399

As raised by @nerolation in research post https://ethresear.ch/t/max-blobs-flag-economic-perspective/21798

@dataalways https://x.com/Data_Always/status/1902697542086721671

And also in the research paper https://arxiv.org/pdf/2502.12966

Blob inclusion is sub-optimal for the builder not taking into account priority fee and packing

Changes

• Solve the 0/1 Knapsack problem to produce optimal blob packing; taking into account tx premium gas (which isn't paid on blobs perse)

Examples from mainnet

6 blobs -> 5 blobs -> 4 blobs; with premium fees to block producer going up x8.8 times

Switch to higher paying blobs

Switch to better paying blobs even if count drops (limited space to include blobs)

Better paying 3 blobs selected over 6 blobs; which causes count to drop to 3 (as max is 6, so the 6 blob tx can't be included as would take to 9)

Types of changes

What types of changes does your code introduce?

• Optimization

Testing

Requires testing

• Yes
• No

If yes, did you write tests?

• Yes
• No