Actual zero or any-value crossing

CROSSINGS returns the crossings of a given value in a vector. Note that it does not return "occurrences" of that value, unless the value is actually crossed.

When there is a value-crossing but not occurrence of that value, the user can determine if they want the crossing to be assigned either
- to the previous element of the input vector (option 'pre'), or
- to the following element (option 'fol'), or
- to that element whose value is closer to the value whose crossings are sought (option 'dis'), or
- to a value calculated after interpolation between the two (preceding and following) elements (option 'int').

Occurrences of the value whose crossings are sought at the beginning or end of the input vector are ignored (since no
actual crossings occur).

Also consecutive occurrences of the value whose crossings are sought, are assigned to the last occurrence before the crossing (see last example).

EXAMPLES
1. Assume that VALUE=0, X(4)=-2 and X(5)=3. Then
- 'dis' will assign the 0-crossing to n=4,
- 'pre' will assign the 0-crossing to n=4,
- 'fol' will assign the 0-crossing to n=5 and
- 'int' will assign the 0-crossing to t=4.4.

2. Assume that the VALUE = 0.
- X = [1 -1 0 -1]: CROSSING will return the 0-crossing from 1 to -1.
- The sequence -1, 0, -1 is not considered a crossing (0 is only reached but not crossed).
- X = [-1 0 0 -1]: CROSSING will not return any 0-crossings (0 is only reached but not crossed).
- X = [1 0 0 -1] : CROSSING will return only one 0-crossing at element 3 (from 0 to -1).

Although the code is checked, please send any overlookings or bugs.

It should work also when using previous versions.