Fix standalone award eligibility to send rich project data matching filtering pass

The Re-evaluate button was producing fewer eligible results because
the standalone job sent minimal project data (title, description, tags)
while the integrated filtering pass sent full data (files, team, institution).

- Fetch rich project data in award-eligibility-job (files, team, institution, etc.)
- Relax AI prompt to be inclusive like the integrated pass — strong primary
  criterion fit is sufficient, don't require all dimensions above 0.5

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

src/server/services/ai-award-eligibility.ts

@@ -63,7 +63,9 @@ quality_score is a 0-100 integer measuring how well the project fits the award c
 
 ## Guidelines
 - Base evaluation only on provided data — do not infer missing information
-- eligible=true only when ALL required dimensions score above 0.5
+- Be inclusive: if the project reasonably fits the criteria, mark eligible=true
+- eligible=true when the project's location, category, or topic aligns with the award criteria
+- Do not require perfection across all dimensions — strong fit in the primary criterion (e.g. geographic) is sufficient
 - confidence reflects how clearly the data supports the determination
 - No personal identifiers in reasoning`
 
@@ -90,7 +92,7 @@ interface ProjectForEligibility {
   institution?: string | null
   foundedAt?: Date | null
   wantsMentorship?: boolean
-  submissionSource?: SubmissionSource
+  submissionSource?: SubmissionSource | string
   submittedAt?: Date | null
   _count?: {
     teamMembers?: number