rewrite description of anatomy.cpp

[MikeL270]

Rewrote the description of anatomy.cpp

[TApplencourt]

What about:

At line 2 we [code]#{hash}include# the SYCL header file, which provides all of
the SYCL features that will be used.

At line 11 we instantiate a <<queue>> with the [code]#in_order# property.
A <<queue>> is bound to a single <<device>>; if you do not specify a device by
passing a selector to the  <<queue>> constructor, the SYCL runtime will choose one
automatically.
A <<queue>> executes the commands submitted to it on its associated <<device>>.
In our case, because of the [code]#in_order# propert, commands are executed in a first in, first out (FIFO) order.
The commands submitted to [code]#myQueue# execute asynchronously with respect to
the host code.

At line 14 we use [code]#malloc_device# to allocate memory on the <<device>>
associated with our <<queue>>, storing a <<usm, unified shared memory (USM)>>
pointer to it in [code]#dataDevice#. This memory is accessible only on the
device.

On lines 17 to 20, [code]#parallel_for# does three conceptual things:

  * the <<sycl-kernel-function>> (defined here as a lambda), passed as the
    second argument to [code]#parallel_for#, is compiled by a
    <<device-compiler>> into a <<kernel>> that can run on the <<device>>
    associated with [code]#myQueue#;
  * a command invoking that <<kernel>> is submitted to [code]#myQueue#, and
    executes asynchronously with respect to the host;
  * when the command runs, the range from 0 to 1023 is decomposed into
    <<work-item,work-items>>, and the <<sycl-kernel-function>> is invoked once
    for each <<work-item>> on the device.

The different <<queue>> member functions used to invoke a <<sycl-kernel-function>> can be found in <<subsec:invokingkernels>>.

At line 23 we use the [code]#copy# member function to submit a copy commands
that copies the results from the device allocation back to host memory.

At line 25 we call [code]#wait# to block the host until all commands submitted
to [code]#myQueue# (the [code]#parallel_for# and the [code]#copy#) have
completed.

At lines 28 to 29 we print the host memory, which now contains a copy of the
device memory.

At line 32, the device allocation [code]#dataDevice# is released with
[code]#free#.

Try to clarify what // for is doing a little clearer (not sure if it's...)