fix(ci): Avoid release artifact collision (#915)

## 📝 Summary

Separate jobs/steps are writing the same rbuilder artifact. Previously
the end result was a linux binary. And lately, it is a macOS binary.

This PR separates artifact writes by OS and arch, in `*.tar.gz` format.

polish constants (#914)

## 📝 Summary

Increased ORDER_JOURNAL_CHANNEL_SIZE x 10 since it would get filled in
busy slots.
Removed 2 unused constants.

## ✅ I have completed the following steps:

* [X] Run `make lint`
* [X] Run `make test`
* [ ] Added tests (if applicable)

Better replacement latancey (#908)

## 📝 Summary

Increassed the ORDERS_CONSUMED_PER_BATCH from 1024 to 4096.
Implemented cancellation in orders queued for simlulation.

## ✅ I have completed the following steps:

* [X] Run `make lint`
* [X] Run `make test`
* [ ] Added tests (if applicable)

---------

Co-authored-by: claude[bot] <209825114+claude[bot]@users.noreply.github.com>

More robust wait for block. (#899)

## 📝 Summary

Some new bug in reth (between 1.9.3 and 1.11.3) generated a rece
condition that would give the last block but then when asked for it it's
not there.
Now wait_for_block_header waits until the deadline no matter the errors
it gets.

## ✅ I have completed the following steps:

* [X] Run `make lint`
* [X] Run `make test`
* [ ] Added tests (if applicable)

---------

Co-authored-by: claude[bot] <209825114+claude[bot]@users.noreply.github.com>

Builder name always needed on rbuilder-operator (#897)

## 📝 Summary

Builder name always needed on rbuilder-operator


## ✅ I have completed the following steps:

* [X] Run `make lint`
* [X] Run `make test`
* [ ] Added tests (if applicable)

Builder server name on TBV stream (#896)

## 📝 Summary

Builder server name on TBV stream

## ✅ I have completed the following steps:

* [X] Run `make lint`
* [X] Run `make test`
* [ ] Added tests (if applicable)

feat: update both optimistic submission methods to use adjustment dat…

…a v3 (#860)

## 📝 Summary

Unify optimistic v1 and v3 submissions to use BidAdjustmentDataV3
format, removing BidAdjustmentDataV1 and intermediate conversion
methods.

## 💡 Motivation and Context

For payout transactions, gas_used may differ from gas_limit.
BidAdjustmentDataV3(https://docs.ultrasound.money/builders/optimistic-v3)
format relaxes the gas_limit == gas_used assumption by explicitly
providing placeholder_gas_used, allowing the relay to use the actual gas
consumed.

In addition to optimistic v3 submissions, there already is a support for
using the BidAdjustmentDataV3 with optimistic v1 submissions. Currently
in rbuilder codebase, v1 submissions used BidAdjustmentDataV1 while v3
submissions used BidAdjustmentDataV2. Changing those submission types to
use v3 data allows relays to use the actual gas consumed, preventing
adjustment failures when gas_used differs from gas_limit, and will
increase builder adjustment revenues.

Therefore we can simplify the codebase by:
- Removing BidAdjustmentDataV1 entirely
- Removing the intermediate BidAdjustmentData struct and conversion
methods
- Using BidAdjustmentDataV3 directly for both optimistic submission
methods

---

## ✅ I have completed the following steps:

* [✅ ] Run `make lint`
* [✅  ] Run `make test`
* [ ] Added tests (if applicable)

cfg for new experimental root hash (#890)

## 📝 Summary

Read the tittle.

## 💡 Motivation and Context

Tried to configure it via telepathy but didn't work.

## ✅ I have completed the following steps:

* [X] Run `make lint`
* [X] Run `make test`
* [ ] Added tests (if applicable)

---------

Co-authored-by: claude[bot] <209825114+claude[bot]@users.noreply.github.com>

Fix Flashbots relay endpoint (#888)

## 📝 Summary

Previously included endpoint was internal without stability guarantees,
switch to public-facing one.

## 💡 Motivation and Context

<!--- (Optional) Why is this change required? What problem does it
solve? Remove this section if not applicable. -->

---

## ✅ I have completed the following steps:

* [ ] Run `make lint`
* [ ] Run `make test`
* [ ] Added tests (if applicable)

Deduplicate payload events by payload attributes only (#880)

Store only PayloadAttributesEvent in recently_sent_data instead of the
full MevBoostSlotData, so deduplication compares only payload
attributes. Remove PartialEq/Eq and the derivative dependency from
MevBoostSlotData since it no longer needs to be comparable.

## 📝 Summary

<!--- A general summary of your changes -->

## 💡 Motivation and Context

<!--- (Optional) Why is this change required? What problem does it
solve? Remove this section if not applicable. -->

---

## ✅ I have completed the following steps:

* [ ] Run `make lint`
* [ ] Run `make test`
* [ ] Added tests (if applicable)