[Animal] Wire per-additive dose fields into methane mitigation calculation

[matthew7838]

Feeds the dose of the selected methane mitigation additive into the enteric methane equations, reading it from that additive's own input field instead of a separate generic field.

Context

Issue(s) closed by this pull request: closes #3030

What

• AnimalConfig gains _METHANE_MITIGATION_DOSE_FIELDS, mapping each methane_mitigation_method to the input field holding its dose.
• AnimalConfig.set_from_data() resolves methane_mitigation_additive_amount through that mapping. "None" and any unrecognized method resolve to 0.0.
• Removes the methane_mitigation_additive_amount input from default.json, example_freestall_animal.json, example_open_lot_animal.json, and the herd manager test fixture.
• Removes the five animal_cross_validation.json blocks that existed to keep the generic field in sync with the per-additive fields.

Why

The animal config carried two overlapping sets of inputs: a generic methane_mitigation_additive_amount, which was the only one Python read, and the four named per-additive fields (3-NOP_additive_amount, monensin_additive_amount, essential_oils_additive_amount, seaweed_additive_amount), which had descriptions and recommended min/max ranges but were never read.

How

set_from_data() looks up the selected method in _METHANE_MITIGATION_DOSE_FIELDS and reads that field. A method with no entry in the mapping (including "None") yields 0.0, which is what the mitigation equations already fall through to.

With the dose now derived rather than user-supplied, the generic input and the five cross-validation rules that policed it are redundant and are deleted. Every one of those rules aliased animal.methane_mitigation.methane_mitigation_additive_amount, so they had to go with the field.

Test plan

• Run with updated unit tests
• Follow issue reproduction steps:
  • Open input/data/animal/example_freestall_animal.json.
  • Set methane_mitigation_method to "3-NOP" and 3-NOP_additive_amount to a non-default value such as 90.
  • Run a simulation and inspect the methane yield reduction — the 3-NOP dose used in the equation should now be 90.

Input Changes

• Deleted animal.methane_mitigation.methane_mitigation_additive_amount from RUFAS/input/metadata/properties/default.json. The dose is now derived from methane_mitigation_method, not supplied by the user.
• Deleted the same key from input/data/animal/example_freestall_animal.json and input/data/animal/example_open_lot_animal.json.
• Deleted five cross-validation blocks from input/metadata/cross_validation/animal_cross_validation.json ("When methane mitigation method is None / 3-NOP / monensin / essential_oils / seaweed_additive, ...").
• No new inputs. The four per-additive fields are unchanged, including their descriptions, defaults, and min/max ranges.
• Existing animal config files that still contain methane_mitigation_additive_amount continue to load; the key is ignored.

Output Changes

• N/A

Filter

No output changes, so no filter is needed.

[github-actions]

Current Coverage: 99%

Mypy errors on methane-mitigation-additive-dose branch: 1136
Mypy errors on dev branch: 1136
No difference in error counts

[github-actions]

🚨 Unauthorized changes detected in protected files. Please remove these changes if they are not intended.

[github-actions]

Current Coverage: 99%

Mypy errors on methane-mitigation-additive-dose branch: 1136
Mypy errors on dev branch: 1136
No difference in error counts

[github-actions]

🚨 Unauthorized changes detected in protected files. Please remove these changes if they are not intended.

[github-actions]

Current Coverage: 99%

Mypy errors on methane-mitigation-additive-dose branch: 1136
Mypy errors on dev branch: 1136
No difference in error counts

[github-actions]

🚨 Unauthorized changes detected in protected files. Please remove these changes if they are not intended.