Policy quality rating treats 'Unclear' rating for Measurable as 'Yes' instead of 'No' (requires update for 1000 Cities Challenge)

[carlhiggs]

Describe the bug
When rating policy for our previous 25 city study, we treated 'unclear' as evidence taking a 'benefit of the doubt' approach to evaluating measurability. We decided to revise that approach in the 1000 Cities challenge (if 'Unclear', then it is not measurable; which is more intuitive). However, when validating manual quality scoring conducted by @mdlowe it was found that the old method was still being used.

In addition, the policy "Transport and planning combined in one government department" was counted towards quality score, but should be omitted from this.

To Reproduce
Using completed policy checklists, the scoring can be checked:
from subprocesses._utils import get_policy_presence_quality_score_dictionary

xlsx = './data/policy_review/2024-02-28  - Completed policy checklists/manual_gohsc-policy-indicator-checklist_Final (1).xlsx'
get_policy_presence_quality_score_dictionary(xlsx)
# {'presence': {'numerator': 26, 'denominator': 32}, 'quality': {'numerator': 20.5, 'denominator': 62}}
## This is similar to score calculated independently by Melanie shared on 28 Feb; in the manual version, quality scored 19.

By making the above changes, the policy score is corrected

>>> get_policy_presence_quality_score_dictionary(xlsx)
{'presence': {'numerator': 26, 'denominator': 32}, 'quality': {'numerator': 19.0, 'denominator': 62}}