There's no other evidence, just a 19th-century historian's arbitrary
emendation of a 17th-century historian's reading of a name in a single
charter (from Gandrea to Gundrea, though arguably an educated guess).
Apart from this unhelpful interference by Arbois de Jubainville with the
spelling of the name, the question raises an interesting issue about the
handling of anomalous sources. Since this must be a case-by-case
problem, it is not often discussed in general.
An interesting example occurs with dating the marriage of Stephen
Henry's son Thibaud IV to Mathilde of Sponheim. Historians usually opt
for 1126, despite a charter dated 1123 in which Mathilde is named by
Thibaud as his wife. The dating to 1126 comes from a contemporary Vita
of St Norbert, founder of the Premonstratensian order, who is said to
have travelled from a meeting with Thibaud (presumably in Troyes) to
Regensburg, where his companions negotiated the marriage, then continued
on to Rome, where we know he had arrived by February 1126. Subsequently
Norbert is said to have been present when the bride did not show up for
the wedding, and he was asked by Thibaud to find out what was causing
the delay. However, the chronology in the Vita is plainly confused,
since according to its account Norbert went from Rome back to Germany
and reached Würzburg by Easter (11 April) 1126 after learning on the way
that he was to be made archbishop of Magdeburg. His election was
formalised in Speyer at the end of June. So in order to believe that he
joined Thibaud for the planned wedding between 11 April and late June
1126, we have to believe that he absented himself from imperial and
Church business for over a month and dashed 600 kms across to France for
a wedding, and moreover that as a prospective German archbishop he was
asked to undertake an errand for a French count as "Father" Norbert. Not
very plausible, but for many historians enough to set aside the evidence
of an otherwise unobjectionably dated charter.
Peter Stewart