Add 6 more February saints/feasts previously never added

Purification of the BVM (Feb 2), St. Andrew Corsini (Feb 4), St. John
of Matha (Feb 8), Apparition of the Immaculate BVM at Lourdes (Feb 11),
Ss. Seven Founders of the Servite Order (Feb 12), St. Gabriel of Our
Lady of Sorrows (Feb 27) -- all had real, complete Matins content in
the reference engine and no exclusion comment; simply never reached
during the earlier month-by-month sanctoral pulls. Continuing the same
sweep as January's St. Francis de Sales/St. Martina/St. John Bosco.

Test update (not a regression): St. Andrew Corsini transfers onto Feb 5
in years where Sexagesima falls on his own Feb 4, outranking St. Agatha
when it does -- february-sanctoral.test.ts's proof year for Agatha
moved from 2029 (now a collision year) to 2025 (clean).

npm test (365 tests) and tsc --noEmit both pass.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

tests/calendar/february-sanctoral.test.ts

@@ -22,7 +22,12 @@ describe('February sanctoral pull (first pass)', () => {
     // reference, not unique text -- confirmed identical to collect-c6.yml's
     // own template; wired via collectCommon (2026-08), postdating this
     // saint's own original authoring, which is why this used to be missing.
-    const day = resolveDay('2029-02-05'); // St. Agatha
+    // 2029 was the original proof year, but St. Andrew Corsini (added
+    // 2026-08 -- see st-andrew-corsini.yml) transfers onto Feb 5 in years
+    // where Sexagesima falls on his own Feb 4, and outranks Agatha
+    // (Duplex vs. Semiduplex) when it does -- 2029 is now one of those
+    // years. 2025 has no such collision.
+    const day = resolveDay('2025-02-05'); // St. Agatha
     expect(day.winner).toEqual({
       kind: 'sanctoral',
       id: 'st-agatha',