[scanner] Don't give the Utf8 scanner a 0-terminated byte sequence

Not having to do the read-allocate-copy dance for files to add a 0-byte
at the end results in these changes when using the CFE to compile
(a fixed version of) the CFE:

```
msec task-clock:u: -1.7356% +/- 0.2164% (-73.16 +/- 9.12)
page-faults:u: -2.6957% +/- 0.0111% (-2914.83 +/- 12.00)
cycles:u: -1.7128% +/- 0.2223% (-297927979.70 +/- 38660477.01)
instructions:u: -1.6814% +/- 0.0002% (-361315766.86 +/- 36853.71)
branch-misses:u: -3.3289% +/- 0.9669% (-2153126.00 +/- 625370.97)
seconds time elapsed: -1.7372% +/- 0.2154% (-0.07 +/- 0.01)
seconds user: -1.5998% +/- 0.2740% (-0.06 +/- 0.01)
seconds sys: -4.1451% +/- 2.9801% (-0.01 +/- 0.01)
Scavenge(   new space) goes from 62 to 61
```

TEST=Existing test coverage.

Change-Id: I8e182bcee39839f6ed1e658c30c85c40ecf0b259
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/385722
Reviewed-by: Johnni Winther <johnniwinther@google.com>
Reviewed-by: Morgan :) <davidmorgan@google.com>
Reviewed-by: Daco Harkes <dacoharkes@google.com>
Reviewed-by: Srujan Gaddam <srujzs@google.com>
Commit-Queue: Jens Johansen <jensj@google.com>
Reviewed-by: Mayank Patke <fishythefish@google.com>
Reviewed-by: Brian Wilkerson <brianwilkerson@google.com>