chore: remove deprecated senate/Triumvirate dead code

[cyclone-pro]

Bug

No existing issue — senate/Triumvirate pallet was silently removed from the subtensor
runtime. The orphaned code remained with TODO comments questioning its status:

# TODO: is this all deprecated? Didn't subtensor senate stuff get removed?

Description of the Change

Removes all dead code related to the senate/Triumvirate voting system that was left
behind after the pallet was dropped from the subtensor runtime.

Files removed or cleaned up:

• bittensor/core/chain_data/proposal_vote_data.py — deleted entirely
• bittensor/core/subtensor.py — removed get_vote_data() method and import
• bittensor/core/async_subtensor.py — removed async get_vote_data() and import
• bittensor/core/chain_data/__init__.py — removed ProposalVoteData export
• bittensor/utils/easy_imports.py — removed ProposalVoteData from public API
• bittensor/extras/subtensor_api/chain.py — removed get_vote_data binding
• tests/unit_tests/test_async_subtensor.py — removed 3 senate-related unit tests
• tests/e2e_tests/test_delegate.py — removed dead import (e2e senate tests were
  already commented out)

Evidence that senate was removed from the runtime:

• Proxy enum entries (Senate, Triumvirate, Governance) were already marked
  # deprecated proxy types in chain_data/proxy.py
• All e2e tests for senate voting in test_delegate.py were already commented out
• No senate extrinsics or commands exist anywhere in the codebase
• No mentions of senate in CHANGELOG for v10.x
• The method queried the Triumvirate pallet's Voting storage — a pallet that no
  longer exists on the active chain

Alternate Designs

Considered adding a DeprecationWarning to get_vote_data() instead of deleting it.
Chose full deletion because the underlying pallet no longer exists on-chain — any call
to this method was already silently returning None or raising a substrate error.
Keeping it would give users false confidence that the API is functional.

Possible Drawbacks

Any external code calling subtensor.get_vote_data() will get an AttributeError
after upgrading. However, since the Triumvirate pallet is gone from the runtime, those
calls were already broken at the chain level. The behavioral change is: fail loudly at
import/call time instead of silently returning None.

Verification Process

1. Searched the full codebase for all references to ProposalVoteData, get_vote_data,
   and Triumvirate — confirmed none remain in production code.
2. Ran the full async subtensor unit test suite:
   pytest tests/unit_tests/test_async_subtensor.py -v

Result: 216 passed, 0 failed, 1 warning (the warning is pre-existing and unrelated).

Release Notes

Removed the defunct senate/Triumvirate voting API (get_vote_data, ProposalVoteData)
that was no longer functional after the senate pallet was removed from the subtensor runtime.

Branch Acknowledgement

[x] I am acknowledging that I am opening this branch against staging

[cyclone-pro]

CC: @basfroman - would appreciate a review when you get a chance.
This removes the orphaned senate/Triumvirate code that was left
after the pallet was dropped from subtensor runtime.

[thewhaleking]

I'm not going to approve or request changes, I'll leave that decision up to @basfroman, but with that being said, I err on the side of not removing dead code in Bittensor until it's "stone dead", because dead things in this ecosystem tend to crop back up with some regularity.

[basfroman]

I'm not going to approve or request changes, I'll leave that decision up to @basfroman, but with that being said, I err on the side of not removing dead code in Bittensor until it's "stone dead", because dead things in this ecosystem tend to crop back up with some regularity.

I completely agree with you. We have already discussed the fate of this code. The decision was to leave it as is until the pallets are updated. In reality, this code is effectively dead. The core of the pallets is slated to be reimplemented within Subtensor; once that happens, we will achieve consistency and update our own components as well.
@cyclone-pro, we really appreciate you for this contribution, but in this particular case, I'd propose closing this PR until the Subtensor pallets have been updated