Switch to #[classmethod] for SpendBundle::py_aggregate

[matt-o-how]

This will allow derived classes to return themselves rather than the SpendBundle parent class

[coveralls-official]

Pull Request Test Coverage Report for Build 10882840971

Warning: This coverage report may be inaccurate.

This pull request's base commit is no longer the HEAD commit of its target branch. This means it includes changes from outside the original pull request, including, potentially, unrelated coverage changes.

• For more information on this, see Tracking coverage changes with pull request builds.
• To avoid this issue with future PRs, see these Recommended CI Configurations.
• For a quick fix, rebase this PR at GitHub. Your next report should be accurate.

Details

• 120 of 169 (71.01%) changed or added relevant lines in 10 files are covered.
• 15 unchanged lines in 6 files lost coverage.
• Overall coverage decreased (-0.02%) to 83.863%

---

Changes Missing Coverage	Covered Lines	Changed/Added Lines	%
crates/chia-bls/src/gtelement.rs	0	5	0.0%
crates/chia-bls/src/public_key.rs	8	13	61.54%
crates/chia-bls/src/secret_key.rs	8	13	61.54%
crates/chia-bls/src/signature.rs	8	13	61.54%
crates/chia-protocol/src/coin_spend.rs	0	5	0.0%
crates/chia-protocol/src/coin.rs	0	6	0.0%
crates/chia-protocol/src/program.rs	0	6	0.0%
crates/chia-consensus/src/gen/owned_conditions.rs	0	12	0.0%

Files with Coverage Reduction	New Missed Lines	%
crates/chia-consensus/src/gen/owned_conditions.rs	1	83.91%
crates/chia-consensus/src/spendbundle_validation.rs	1	99.71%
crates/chia-consensus/src/gen/make_aggsig_final_message.rs	1	98.59%
crates/clvm-derive/src/from_clvm.rs	2	98.78%
crates/clvm-derive/src/to_clvm.rs	3	98.1%
crates/clvm-derive/src/parser/attributes.rs	7	86.67%

Totals
Change from base Build 10422144649:	-0.02%
Covered Lines:	12483
Relevant Lines:	14885

---

💛 - Coveralls

[Rigidity]

Did you change the line endings in these files? The diff seems way too large

[arvidn]

I would have thought you'd need to use the passed-in cls type eomshow

[arvidn]

we should try going back to invoking the derived class with all the fields of the class itself, as if it is a dataclass

[arvidn]

I think there are a few simplification opportunities in the error propagation, and I think the type stubs can be improved

[arvidn]

looking good. I thought you used to have a python test as well, to ensure it works

[arvidn]

look good, you just missed one place that sill uses with_gil()