Upgrade to Tika 1.10

Release 1.10 - 8/1/2015

  * Tika Config XML can now be used to create composite detectors,
    and exclude detectors that DefaultDetector would otherwise
    have used. This brings support in-line with Parsers. (TIKA-1702)

  * Reverted to legacy sort order of parsers that was 
    mistakenly reversed in Tika 1.9 (TIKA-1689).

  * Upgrade to POI 3.13-beta1 (TIKA-1667).

  * Upgrade to PDFBox 1.8.10 (TIKA-1588).

  * MimeTypes now tries to find a registered type with and 
    without parameters (TIKA-1692).

  * Added more robust error handling for encoding detection
    of .MSG files (TIKA-1238).

  * Fixed bug in Tika's use of the Jackcess parser that 
    prevented reading of v97 Access files (TIKA-1681).

  * Upgrade xerial.org's sqlite-jdbc to 3.8.10.1. NOTE: 
    as of Tika 1.9, this jar is "provided." Make sure 
    to upgrade your provided jar! (TIKA-1687).

  * Add header/footer extraction to xls (via Aeham Abushwashi)
    (TIKA-1400).

  * Drop the source file name from the embedded file path in
    RecursiveParserWrapper's "X-TIKA:embedded_resource_path" 
    (TIKA-1673).

  * Upgraded to Java 7 (TIKA-1536).

  * Non-standards compliant emails are now correctly detected
    as message/rfc822 (TIKA-1602).

  * Added parser for MS Access files via Jackcess. Many thanks 
    to Health Market Science, Brian O'Neill and James Ahlborn 
    for relicensing Jackcess to Apache v2! (TIKA-1601)

  * GDALParser now correctly sets "nitf" as a supported 
    MediaType (TIKA-1664).

  * Added DigestingParser to calculate digest hashes 
    and record them in metadata. Integrated with
    tika-app and tika-server (TIKA-1663).

  * Fixed ZipContainerDetector to detect all IPA files
    (TIKA-1659).