Add v2_a11 describability axis to filter_score

Add an 11th composite axis 'v2_a11_describability' scored 0-3 measuring whether the proposal's *change_type + target_files + proposed_diff* tuple is concretely describable in <=2 sentences without invoking abstract nouns ('clarity', 'robustness', 'better'). Implementation: regex-blocklist over proposed_diff for vague-noun stems (clar*, robust*, better, improv*, optim*, enhance*) AND a check that target_files contains >=1 concrete repo path that exists on disk (fs.existsSync). Score = 3 if zero vague-stem hits AND >=1 valid path; 2 if 1 vague hit; 1 if 2 hits or no path; 0 otherwise. Insert axis registration after the v2_a10 block and include in composite_v2 sum.

On the 21 currently-graduated meta proposals, the 847f7e-shape (vague-clarity) proposals score 0-1 while concrete-diff proposals (e.g. the S179 handoff-contract style) score 2-3, producing a 2-3 point spread in composite_v2 that correlates with downstream council_verdict APPROVE rate at >=0.4 Pearson.

Rerun on the 43 historical filter_score rows; if a11 score has |r|<0.2 with council_verdict APPROVE and does not separate the known 7 847f7e-shape rejects from the ROBUST set, the axis is misdesigned and gets reverted.

Rejected by filter_score. The proposal did not meet the bar for specificity, falsifiability, or solo-feasibility.