Almost all LDAP schemas use an attribute (uid or mail) for the user name that is specified to be case insensitive. Unless you are connecting to a LDAP server that uses a different search attribute than the defaults you should leave this as Case insensitive.

Recommendations:

Note:

If you are using an email address attribute for theĀ user search you will want to ask the mail server administrator whether they have case sensitive mailboxes. RFC 2821 Section 2.4 specifies:
The local-part of a mailbox MUST BE treated as case sensitive. Therefore, SMTP implementations MUST take care to preserve the case of mailbox local-parts. Mailbox domains are not case sensitive.
This is in conflict with the case sensitivity rules for the standard LDAP attribute used to hold email addresses (mail) which specifies caseIgnoreIA5Match, e.g. as noted in RFC 2798
( 0.9.2342.19200300.100.1.3
           NAME 'mail'
           EQUALITY caseIgnoreIA5Match
           SUBSTR caseIgnoreIA5SubstringsMatch
           SYNTAX 1.3.6.1.4.1.1466.115.121.1.26{256} )
         
Note: RFC 1274 uses the longer name 'rfc822Mailbox' and syntax OID of 0.9.2342.19200300.100.3.5. All recent LDAP documents and most deployed LDAP implementations refer to this attribute as 'mail' and define the IA5 String syntax using using the OID 1.3.6.1.4.1.1466.115.121.1.26, as is done here.
So when the user search filter is mail={0} it is pointless to select anything other than Case insensitive as the LDAP server will be ignoring case when searching anyway.
If you do need to select a case sensitive strategy because you are searching on a custom attribute you should note that the Case sensitive (email address) strategy applies RFC 2821's rules on case sensitivity (i.e. the local-part is case sensitive and the mailbox domain is case insensitive), thus with that strategy JOE@ACME.COM and JOE@acme.com will be considered the same Jenkins user while joe@acme.com will be considered as a different Jenkins user.

Note: UNIX login names are case sensitive