SeedChangeRequests: two comments name seeded requests in a spelling the code no longer uses

[dfe-lance]

Problem

Two comments in SeedChangeRequests.cs name a seeded request in a spelling the line
beside them no longer uses:

(Reference: Reference(7), Status: RequestStatus.ReadyToSubmit, Pupil: p[5]), // duplicate of SEED006
(Reference: Reference(9), Status: RequestStatus.ReadyToSubmit, Pupil: p[6]), // duplicate of already-submitted SEED008

The references used to be written out in full ("CYPMD_KS4June_SEED007"). They are now
generated, so the reset behind /dev/queues/cleanup-e2e-requests can derive "these rows are
the seeded baseline, keep them" from the same place the rows are written and the two cannot
drift apart. The comments were left as they were.

They are still accurate — Reference(n) produces exactly CYPMD_KS4June_SEED{n:000}, so
the seeded data is unchanged, and Reference(7) is CYPMD_KS4June_SEED007. The problem is
only that a reader now has to map between two names for the same thing to see which row is
being pointed at, and the obvious reading — that the comment refers to something the code no
longer creates — is wrong.

Raised in review.

Fix

Refer to the sibling row the way the line beside it does:

(Reference: Reference(7), Status: RequestStatus.ReadyToSubmit, Pupil: p[5]), // duplicate of Reference(6)
(Reference: Reference(9), Status: RequestStatus.ReadyToSubmit, Pupil: p[6]), // duplicate of already-submitted Reference(8)

Comments only. No behaviour change, and no change to what is seeded.

Acceptance criteria

• Neither comment names a reference in a spelling the code does not use
• The seeded rows are unchanged — still CYPMD_KS4June_SEED001 to 011, same statuses,
  same pupils